九州大学 研究者情報
論文一覧
廣川 佐千男(ひろかわ さちお) データ更新日:2019.06.03

教授 /  情報基盤研究開発センター 学術情報研究部門


原著論文
1. Jun Zeng, Xin He, Yingbo Wu, Sachio Hirokawa, User Behavior Analysis of Location-Based Social Network, 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
Proceedings - 2018 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
, 10.1109/IIAI-AAI.2018.00015, 21-25, 2019.04, [URL], User behavior changes over time under the influence of their activities. We contend that these activities are non-random behavior and have a desire to explore the underlying information behind these changes. In this paper, we analyze user behavior by using the check-in data in Location Based Social Networks (LBSNs), and examine whether they have the features of trend, periodicity and surprise or not. We explore some dynamics behaviors of people through their check-in times, and divide time into annually, monthly and even weekly analysis to find out the pattern of their behavior. Eventually, we found the check-in data do exhibit these three features by analyzing them deeply. The analytical work lays the foundation for the further recommendation research..
2. Sachio Hirokawa, Kiyota Hashimoto, Simplicity of Positive Reviews and Diversity of Negative Reviews in Hotel Reputation, 2018 International Joint Symposium on Artificial Intelligence and Natural Language Processing, iSAI-NLP 2018
2018 International Joint Symposium on Artificial Intelligence and Natural Language Processing, iSAI-NLP 2018 - Proceedings
, 10.1109/iSAI-NLP.2018.8692973, 2019.04, [URL], User's review on products and services is valuable information for both users and providers. The present paper conducted a polarity estimation of 73,589 reviews on hotels in Europe. Users rated one to five points for seven aspects (Value, Rooms, Location, Cleanliness, Check-in, Service, Business, Overall). In this paper, we predicted the polarity (positive/negative) of each aspect by using a machine learning method, SVM (Support Vector Machine), and feature selection, with more than 4 points being positive and less than 3 being negative. As a result, positive reviews with respect to six aspects, other than Business, were able to achieve 74
%
prediction performance (F-measure) with only 20 feature words. On the other hand, for negative reviews, optimal prediction performance could not be obtained unless almost all words were used, and on average F-measure was only 27%. The results indicate that positive reviews are simple, meanwhile negative reviews are diverse and hard to predict mechanically..
3. Kahori Ogashiwa, Toru Sugihara, Kumiko Kanekawa, Yuki Kitanaka, Kazuhisa Noguchi, Soichiro Aihara, Masao Mori, Sachio Hirokawa, Quality Assurance in Education through the Diploma Policy and President's Message - An Analysis Focused on Local Community, 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
Proceedings - 2018 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
, 10.1109/IIAI-AAI.2018.00203, 964-965, 2019.04, [URL], Universities play an important role as centers of local communities. In recent years, new faculties and graduate schools have been established for students to learn about solving problems through local community involvement. In this study, we conduct an inter-university comparison and analyze the common points and differences between eight targeted universities. These include national universities (nine departments) that use the word 'community' ('chi-i-ki' in Japanese) in the name of a department. We use the term frequency-inverse document frequency (tf-idf), which is one of the popular indexes in text mining, and apply it to two documents, i.e., the diploma policy (DP) and the president's message (PM). We extracted two groups from the target universities. One group had relatively high tf-idf value in terms of the word 'community' across the two documents. The other group had a relatively low tf-idf value in their DP. Furthermore, the word 'community' does not appear in the PM. All universities have established faculties in order to connect with local communities, although the strategies of each university demonstrate different priorities from each other..
4. Sachio Hirokawa, Message from General Chair, 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
Proceedings - 2018 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
, 10.1109/IIAI-AAI.2018.00005, XXIV, 2019.04, [URL].
5. Takahiko Suzuki, Tssukasa Kamimasu, Tetsuya Nakatoh, Sachio Hirokawa, Identification of Unnatural Subsets in Statistical Data, 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
Proceedings - 2018 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
, 10.1109/IIAI-AAI.2018.00024, 74-80, 2019.04, [URL], Benford's law is an observation on the frequency distribution of first significant digits in natural numerical data. We can measure the unnaturalness of the data by evaluating estrangement of the frequency distribution of leading digits of the data in relation to the Benford's distribution. However, we cannot identify the unnatural part of the data precisely. In this study, we focus on the fact that statistical data is generally provided in tabular form. We specify a subset of the target data by using the item names of rows and columns that define each cell of the table or words appearing in the table title. By measuring the degree of divergence of the subset from Benford's distribution, we can identify unnatural subsets. We apply this method to agriculture-related data from China Statistical Yearbook and succeeded to identify unnatural subsets..
6. Masaru Taga, Toshihiro Onishi, Sachio Hirokawa, Automated Evaluation of Students Comments Regarding Correct Concepts and Misconceptions of Convex Lenses, 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
Proceedings - 2018 7th International Congress on Advanced Applied Informatics, IIAI-AAI 2018
, 10.1109/IIAI-AAI.2018.00059, 273-277, 2019.04, [URL], It is important for teachers to understand whether students have learned correct concepts or misconceptions. To identify correct concepts and misconceptions, we attempted to analyze the characteristics of students' concept descriptions using machine learning. It has been mentioned that university students retain the misconception that 'light refracts at the center line of a convex lens' due to the influence of the learning construction method in junior high school. Therefore, the descriptions from 104 Japanese university third-graders (36 from the faculty of social studies and 68 from the faculty of science and engineering) were analyzed by text mining using a support vector machine, and their feature words were identified. As a result, it was possible to classify the characteristics of the descriptions as a correct concept or misconception based on the feature words..
7. Sachio Hirokawa, Key attribute for predicting student academic performance, 10th International Conference on Education Technology and Computers, ICETC 2018
Proceedings of the 10th International Conference on Education Technology and Computers, ICETC 2018
, 10.1145/3290511.3290576, 308-313, 2018.10, [URL], Predicting student final score from student's attributes is an important issue of learning analytic. Not only to achieve high prediction performance but also to identifying the key attributes is an important research theme. This paper evaluated exhaustively the prediction performance based on all possible combinations of four types of attributes - behavioral features, demographic features, academic background, and parent participation. The behavioral features are given as numerical data. But, we represented them as pair of an attribute name and the value. This vectorization yields 417 dimensional data, while naively represented data has 68 dimension. By applyig support vector machine and feature selection, we obtained the optimal prediction performance, with respect to feature selection, with accuracy 0.8096 and F-measure 0.7726. We confirmed that the behavioral feature is so crucial that the accuracy reaches 0.7905 without other features except behavioral feature. The combination of behavior feature and demographic feature gained F-measure 0.7662..
8. Yoshiki Mashima, Sachio Hirokawa, Kazuhiro Takeuchi, Ties between mined structural patterns in program and their identifier names, 7th International Symposium on Integrated Uncertainty in Knowledge Modelling and Decision Making, IUKM 2019
Integrated Uncertainty in Knowledge Modelling and Decision Making - 7th International Symposium, IUKM 2019, Proceedings
, 10.1007/978-3-030-14815-7_28, 335-346, 2019.01, [URL], Identifier names in readable and maintainable source codes are always descriptive. These names are given based on the implicit knowledge of experienced programmers. In this paper, we propose a structural pattern mining method based on support vector machines (SVM) for source codes. We extract 1,000 method names in object-oriented source codes collected from online software repositories and create 1,000 datasets labeled by positive and negative class. The structural features used for the input feature vectors to the SVM learning are designed for representing partial characteristics in the abstract syntax tree (AST) parsed from a source code. Applying this method, we made an F1 score list of the 1,000 method names, which shows the degree of patterning of each name, by using our structural features. From the list, we confirmed structural patterns were strongly associated with specific method names. A qualitative evaluation of method names was also conducted by mapping the structural feature vector of each program example to the two-dimensional plane in the same way as a previous major study. From the evaluation, we confirmed that the contrasting structure among the programs corresponds to the names given to programs. Furthermore, we show examples of visualization of structural patterns using structural features extracted by feature selection..
9. Sachio Hirokawa, Good students look back previous pages, 26th International Conference on Computers in Education, ICCE 2018
ICCE 2018 - 26th International Conference on Computers in Education, Workshop Proceedings
, 457-466, 2018.11, Educational institutions have many expectations for the use of E-book. The top expectation is to evaluate and to improve the education system based on the accumulated learning activity log data. This paper applied machine learning to predict the learner's final score from e-Book browsing logs. The present paper evaluated the prediction performance of the good students with the final grade of 80 or more from their learning access logs. An experimental evaluation revealed that the prediction performance (accuracy) was only 64% if we use only the accessed page information. However, the accuracy was improved to 89% when consecutive browsing page transition information was used. Furthermore, it was confirmed that returning to the previous page as a feature of the highest grades student.s..
10. Brendan Flanagan, Sachio Hirokawa, An automatic method to extract online foreign language learner writing error characteristics, International Journal of Distance Education Technologies, 10.4018/IJDET.2018100102, 16, 4, 15-30, 2018.10, [URL], This article contends that the profile of a foreign language learner can contain valuable information about possible problems they will face during the learning process, and could be used to help personalize feedback. A particularly important attribute of a foreign language learner is their native language background as it defines their known language knowledge. Native language identification serves two purposes: to classify a learners' unknown native language; and to identify characteristic features of native language groups that can be analyzed to generate tailored feedback. Fundamentally, this problem can be thought of as the process of identifying characteristic features that represent the application of a learner's native language knowledge in the use of the language that they are learning. In this article, the authors approach the problem of identifying characteristic differences and the classification of native languages from the perspective of 15 automatically predicted writing errors by online language learners..
11. Takanori Yamashita, Naoki Nakashima, Sachio Hirokawa, Classification and feature extraction for text-based drug incident report, 6th International Conference on Bioinformatics and Computational Biology, ICBCB 2018
Proceedings of 2018 6th International Conference on Bioinformatics and Computational Biology, ICBCB 2018
, 10.1145/3194480.3194499, 145-149, 2018.03, [URL], Medical institutions have been constructed incident report system, then accumulating incident data. Incident data compose text-based data and some structured attributes. We considered based on the analysis result with clustering for drug incident report. Firstly, we generated a network of documents and words from the text-based data. Secondly, Louvain method was applied to the network and 11 clusters were generated. We confirmed the contents of each cluster from feature words extracted by TF-IDF. Then, we compare clusters of text-based data with structured attributes and grasp the trend of the incident. This proposed method showed the possibility of clinical support toward reduction incident from text-based data..
12. N. Onimura, T. Yamashita, N. Nakashima, H. Soejima, S. Hirokawa, Machine Learning Support for Template Design of Clinical Notes, Proc. of the Eighth International Conference On Advances in Computing, Electronics and Electrical Technology - CEET 2018, pp.19-24, 2018, 2018.03.
13. Yoshiki Mashima, Takuya Okada, Sachio Hirokawa, Kazuhiro Takeuchi, Predicting Purpose of Program from Superficial Structure, Proc. ASEAN-AI2018 (in press), 2018.03.
14. Kumiko Kanekawa, Tetsuya Nakatoh, Takahiko Suzuki, Sachio Hirokawa, Assessment of Doctoral Supervision of International Students, Proc. of International Conference New Perspectives in Science Education, pp.415-421, 2018, 2018.03.
15. Yao Lin, Kohei Yamaguchi, Tsunenori Mine, Sachio Hirokawa, Is SVM+FS better to satisfy decision by majority?, 3rd International Conference on Soft Computing and Data Mining, SCDM 2018
Recent Advances on Soft Computing and Data Mining - Proceedings of the 3rd International Conference on Soft Computing and Data Mining SCDM 2018
, 10.1007/978-3-319-72550-5_26, 700, 261-271, 2018.01, [URL], Government 2.0 activities have become very attractive and popular. Using the platforms to support the activities, anyone can anytime report issues in a city on the Web and share the reports with other people. Since a variety of reports are posted, officials in the city management section have to give priorities to the reports. However, it is not easy task to judge the importance of the reports since importance judgments vary depending on the officials and consequently the agreement rate becomes low. To remedy the low agreement rate problem of human judgment, it is necessary to create an automatic method to find reports with high priorities. Hirokawa et al. employed the Support Vector Machine (SVM) with word feature selection method (SVM+FS) to detect signs of danger from posted reports because signs of danger is one of high priority issues to be dealt with. However they did not compare the SVM+FS method with other conventional machine learning methods and it is not clear whether or not the SVM+FS method has better performance than the other methods. This paper compared the results of the SVM+FS method with conventional machine learning methods: SVM, Random Forest, and Naïve Bayse with conventional word vectors, an LDA-based document vector, and word embedding by Word2Vec. Experimental results illustrate the validity and effectiveness of the SVM+FS method..
16. Toshiro Minami, Sachio Hirokawa, Yoko Ohura, Kiyota Hashimoto, A part-of-speech-based exploratory text mining of students’ looking-back evaluation, 11th International Symposium on Natural Language Processing, SNLP-2016 and 1st Workshop in Intelligent Informatics and Smart Technology, 2016
Advances in Natural Language Processing, Intelligent Informatics and Smart Technology - Selected Revised Papers from the 11th International Symposium on Natural Language Processing SNLP-2016 and the 1st Workshop in Intelligent Informatics and Smart Technology
, 10.1007/978-3-319-70016-8_6, 61-72, 2018.01, [URL], In our lectures at universities, we observe that the students’ attitudes affects a lot to their achievements. In order to prove this observation based on data, we have been investigating to find effective methods that extract students’ attitudes from lecture data; such as examination score as an index to student’s achievement, attendance and homework data for his/her effort, and answer texts of the term-end questionnaire as information source of attitude. In this chapter, we take another approach to investigate the influences of words used in the answer texts of students on their achievements. We use a machine learning method called Support Vector Machine (SVM), which is a tool to create a model for classifying the given data into two groups by positive and negative training sample data. We apply SVM to the answer texts for analyzing the influences of parts of speech of words to the student’s achievement. Even though adjectives and adverbs are the same in the sense that they modify nouns and verbs, we found that adverbs affects much more than adjectives, as a result. From our experiences so far, we believe that analysis of answers to the evaluations of students toward themselves and lectures are very useful source of finding the students’ attitudes to learning..
17. Takanori Yamashita, Naoya Onimura, Hidehisa Soejima, Naoki Nakashima, Sachio Hirokawa, Graph Clustering System for Text-Based Records in a Clinical Pathway, Studies in Health Technology and Informatics, 245, 649-652, 2018.01, The progressive digitization of medical records has resulted in the accumulation of large amounts of data. Electronic medical data include structured numerical data and unstructured text data. Although text-based medical record processing has been researched, few studies contribute to medical practice. The analysis of unstructured text data can improve medical processes. Hence, this study presents a clustering approach for detecting typical patient's condition from text-based medical record of clinical pathway. In this approach, the sentences in a cluster are merged to generate a "sentence graph" of the cluster after classified feature word by Louvain method. An analysis of real text-based medical records indicates that sentence graphs can represent the medical treatment and patient's condition in a medical process. This method could help the standardization of text-based medical records and the recognition of feature medical processes for improving medical treatment..
18. Chao Zeng, Tetsuya Nakatoh, Sachio Hirokawa, Masanari Eguchi, Text mining of tourism preference in a multilingual site, IEEJ Transactions on Electrical and Electronic Engineering, 10.1002/tee.22841, 2018.01, [URL], There is a huge demand on multilingual tourism information of Japan because of the increasing number of tourists from foreign countries. Most of them may expect typical and stereotyped culture, nature, and modern society of Japan. However, people from different backgrounds, cultures, and languages might expect different aspects of Japan, as well. In this paper, we analyze these kinds of differences as the cultural tourism preference for Japan. We propose a machine-learning-based method to figure out the cultural tourism preference of people of different countries based on comparing the access logs to a multilingual tourism information site in different languages. We focus our discussion on the pages accessed in Thai and Vietnamese languages. Our research result shows that for Thai tourists the characteristic features are the famous places in an area and local specialties, but Vietnamese tourists pay much more attention to facilities and location of hotels. This difference was not observable by naive extraction of keywords and their visualization. This result has been used as a guide to the further creation of content in the tourism information site..
19. Tetsuya Nakatoh, Kenta Nagatani, Kumiko Kanekawa, Takahiko Suzuki, Sachio Hirokawa, Cluster analysis of scientific citation context, 19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017
19th International Conference on Information Integration and Web-Based Applications and Services, iiWAS2017 - Proceedings
, 10.1145/3151759.3151811, 111-115, 2017.12, [URL], Investigation of related research is a very important task for researchers. In recent years, databases of academic papers have been developed, and researchers can search for related research using keywords and so on. However, it is a very time-consuming task to discover appropriate papers exhaustively from a large number of academic papers, classify them, and understand their contents. We have been researching methods that can properly extract papers of related research from an academic paper database. After finding those papers, researchers need to understand the relationship between those papers. We believe that there are several types of relations between papers that appear in citation expressions of related papers. In this paper, automatic classification of citation expressions is performed as the first step in the analysis of citation expressions, and the analysis results of each cluster are reported..
20. Yudai Tanabe, Koki Kagari, Yuki Kitanaka, Kazuhiro Takeuchi, Sachio Hirokawa, Finding key integer values in many features for learners' academic performance prediction, 9th International Conference on Education Technology and Computers, ICETC 2017
Proceedings of the 9th International Conference on Education Technology and Computers, ICETC 2017
, 10.1145/3175536.3175551, 167-171, 2017.12, [URL], In recent years, along with the proliferation of the learning management system (LMS), a large amount of data regarding the interaction between the system and the learners has been accumulated. Correspondingly, various data mining methods have been applied to these data. In order to employ a suitable computational model that is at the core of the data mining method and is not automatically acquired by the mining method itself, it is important to make or find various reasonable hypotheses for target variables. In this paper, we propose a method for analyzing closely the degree to which the explanatory variables represented in integer value contributes to predicting categorical objective variables, such as a learner's academic performance. Specifically, we describe that a decision tree combining support vector machines (SVM) achieves accuracy consistent with existing research, and it contributes further extraction of particular explanatory values from the integer features. Before making a model with SVM, our proposal method expands original features represented by integer value to corresponding binary features. With this expansion of original features, we can identify the key values that closely relate to a learner's academic performance from behavioral features gathered from LMS. Identifying such key values in specific features plays an important role in developing a hypothesis that explains the objective variables, using them as explanatory variables. We believe that closer analysis of these key explanatory values will find latent knowledge that can improve learners' academic abilities..
21. Yuki Kitanaka, Kazuhiro Takeuchi, Sachio Hirokawa, Predicting learning result of learner in e-learning course with feature selection using SVM, 9th International Conference on Education Technology and Computers, ICETC 2017
Proceedings of the 9th International Conference on Education Technology and Computers, ICETC 2017
, 10.1145/3175536.3175567, 122-125, 2017.12, [URL], In recent years, data mining targeting educational data has been widely performed. With the spread of the e-learning system, activities of various learners have been recorded. By analyzing this record, research is being conducted to evaluate the achievement level of the learner and to find hidden problems. In this paper, we compare the existing method and the method by SVM using feature selection for the method of classifying the final result from the learner's activity record. This confirms the effectiveness of the method using feature selection. Next, we confirmed that the click stream which is the activity data in the elearning system is more effective than the learner's profile in classification of grades. In the classification of learners with good grades, the connection records and the number of clicks in the latter period of the learning period are important factors, further the difference in the important features by the grade evaluation was shown..
22. Eisuke Ito, Yuya Honda, Sachio Hirokawa, Empathy factor mining from reader comments of e-manga, Proc. of eKnow2018, IARIA, pp.107-112, 2018, 10.1145/3175536.3175567, 2017.12.
23. Tetsuya Nakatoh, Sachio Hirokawa, Toshiro Minami, Takeshi Nanri, Miho Funamori, Attribute-based quality classification of academic papers, Artificial Life and Robotics, 10.1007/s10015-017-0412-z, 1-6, 2017.11, [URL], Investigating the relevant literature is very important for research activities. However, it is difficult to select the most appropriate and important academic papers from the enormous number of papers published annually. Researchers search paper databases by combining keywords, and then select papers to read using some evaluation measure—often, citation count. However, the citation count of recently published papers tends to be very small because citation count measures accumulated importance. This paper focuses on the possibility of classifying high-quality papers superficially using attributes such as publication year, publisher, and words in the abstract. To examine this idea, we construct classifiers by applying machine-learning algorithms and evaluate these classifiers using cross-validation. The results show that our approach effectively finds high-quality papers..
24. Brendan Flanagan, Sachio Hirokawa, Emiko Kaneko, Emi Izumi, Hiroaki Ogata, A Multi-model SVR Approach to Estimating the CEFR Proficiency Level of Grammar Item Features, 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
Proceedings - 2017 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
, 10.1109/IIAI-AAI.2017.169, 521-526, 2017.11, [URL], Analysis of publicly available language learning corpora can be useful for extracting characteristic features of learners from different proficiency levels. This can then be used to support language learning research and the creation of educational resources. In this paper, we classify the words and parts of speech of transcripts from different speaking proficiency levels found in the NICT-JLE corpus. The characteristic features of learners who have the equivalent spoken proficiency of CEFR levels A1 through to B2 were extracted by analyzing the data with the support vector machine method. In particular, we apply feature selection to find a set of characteristic features that achieve optimal classification performance, which can be used to predict spoken learner proficiency..
25. Jun Zeng, Yinghua Li, Feng Li, Junhao Wen, Sachio Hirokawa, A Point-of-Interest Recommendation Method Using Location Similarity, 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
Proceedings - 2017 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
, 10.1109/IIAI-AAI.2017.122, 436-440, 2017.11, [URL], POI recommendation aims to recommend places which users have not visited before. In this paper, we proposed a POI recommendation method using location similarity, which assumes that people may be interested in the places that are similar with the places that they have been to before. In order to calculate the similarity of locations, we proposed a novel method using time slots. Every two hours can be considered as a time slot. In other words, one day can be segmented into 12 time slots. For each location, the check-in times in each time slot can be collected. These check-in times can form a vector, which can be used to calculate the similarity of two locations. According to the similarity, the score of each unvisited locations can be calculated and sorted. Finally, the POI recommendation can be generated from the top-n unvisited locations. The experiment results show that the proposed method is effective..
26. Toru Sugihara, Soichiro Aihara, Sachio Hirokawa, Takashi Nara, An Analysis of Characteristics of Student-Athletes from Questionnaire by SVM, 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
Proceedings - 2017 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
, 10.1109/IIAI-AAI.2017.215, 163-166, 2017.11, [URL], What sort of care should a university take for student-athletes? To answer the question and to consider the future educational strategy are one of big issues for many universities. The authors created a questionnaire which consists of 77 questions with multiple choice form. We collected the responses from 100 student-athletes and 141 other students. The present paper analyzed the characteristic features of student-athletes. We considered 312 kinds of combination of question items and the response choices as words and the questionnaire record of a student as a document written in those words. Then we applied the text mining method SVM (support vector machine) and feature selection. As the result, we confirmed that we can distinguish student-athletes from other students with 90% accuracy based on 16 characteristic features such as (a) they spend much time on athlete club and not on study, (b) they want to work for economically rich life, (c) they think that it is advantageous to job hunting or graduate school if they have good grades and (d) they have less interests on international perspective in campus life..
27. Kumiko Kanekawa, Tetsuya Nakatoh, Takahiko Suzuki, Sachio Hirokawa, Who is the Last Author of Your Paper?, 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
Proceedings - 2017 6th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2017
, 10.1109/IIAI-AAI.2017.124, 221-224, 2017.11, [URL], Evaluation of researchers is a big issue in institutional research. We propose a method for quantitatively evaluating the stage of young, middle and senior researchers focusing on the role of the last author in co-authored papers. We trace the two time series of the number of published papers and the ratio of the last authored papers among them of each researcher. We conducted experiments on 84 researchers of ICT related graduate school of a university in Japan, and on 50 researchers who published papers in 15 highly evaluated international conferences and 5 international journals. We analyzed 3360 articles in the first case and 13138 articles in the second case. We test three different approaches: cross tables, portfolios and bar graphs..
28. 鈴木聡, 廣川佐千男, ペアプログラミングと反転授業を導入したコンピュータシミュレーション実習における履修者の学習活動の分析, 日本教育工学会論文誌, Vol.41, No.3, pp.245-253, 2017, 2017.10.
29. Sachio Hirokawa, Takahiko Suzuki, Tsunenori Mine, Machine learning is better than human to satisfy decision by majority, 16th IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017
Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017
, 10.1145/3106426.3106520, 694-701, 2017.08, [URL], Government 2.0 activities have become very attractive and popular these days. Using platforms to support the activities, anyone can anytime report issues or complaints in a city with their photographs and geographical information on the Web, and share them with other people. Since a variety of reports are posted, officials in the city management section have to check the importance of each report and sort out their priorities to the reports. However, it is not easy task to judge the importance of the reports. When several officials work on the task, the agreement rate of their judgments is not always high. Even if the task is done by only one official, his/her judgment sometimes varies on a similar report. To remedy this low agreement rate problem of human judgments, we propose a method of detecting signs of danger or unsafe problems described in citizens' reports. The proposed method uses a machine learning technique with word feature selection. Experimental results clearly explain the low agreement rate of human judgments, and illustrate that the proposed machine learning method has much higher performance than human judgments..
30. Takahiko Suzuki, Koki Miyata, Sachio Hirokawa, Difficulty of words and their ambiguity estimated from the result of word sense disambiguation, 11th International Conference on Knowledge, Information and Creativity Support Systems, KICSS 2016
Proceedings - 11th 2016 International Conference on Knowledge, Information and Creativity Support Systems, KICSS 2016
, 10.1109/KICSS.2016.7951414, 2017.06, [URL], When learning a new word in language learning, there are two problems. One is how difficult the word itself is. The second is, in what kind of situation, it will be used. There is a research that defined quantitative ambiguity of words based on the structure of WordNet, then investigated the relationship between the ambiguity and the difficulty level of words. In this paper, we re-define ambiguity of word occurrences in text by using the result of word sense disambiguation technique. We analyze the relationship between the ambiguity of words and their difficulty level. We compare the result with those in the previous research. Utilizing knowledge and training data affect the relationship between the difficulty and ambiguity of words..
31. Yusuke Adachi, Naoya Onimura, Takanori Yamashita, Sachio Hirokawa, Classification of imbalanced documents by feature selection, 2017 International Conference on Compute and Data Analysis, ICCDA 2017
Proceedings of 2017 International Conference on Compute and Data Analysis, ICCDA 2017
, 10.1145/3093241.3093246, Part F130280, 228-232, 2017.05, [URL], We previously worked on category classification problem of reuter's newspaper article using SVM and feature selection. In the study, feature selection by SVM-score [Sakai, Hirokawa, 2012] showed high accuracy. It was also expected to be superior to other standard indicators in case data is imbalanced. This study aimed to show the effectiveness of feature selection by SVM-score in machine learning with imbalanced data. For the reuter's data, F-measure was calculated in the classification experiment of all 13 categories. As a result, feature selection by SVM-score shows high f-measure and precision. In addition, we found feature words of negative example improve the classification performance..
32. Jun Zeng, Feng Li, Brendan Flanagan, Sachio Hirokawa, LTDE
A layout tree based approach for deep page data extraction, IEICE Transactions on Information and Systems, 10.1587/transinf.2016EDP7375, E100D, 5, 1067-1078, 2017.05, [URL], Content extraction from deep Web pages has received great attention in recent years. However, the increasingly complicated HTML structure of Web documents makes it more difficult to recognize the data records by only analyzing the HTML source code. In this paper, we propose a method named LTDE to extract data records from a deepWeb page. Instead of analyzing the HTML source code, LTDE utilizes the visual features of data records in deep Web pages. A Web page is considered as a finite set of visual blocks. The data records are the visual blocks that have similar layout. We also propose a pattern recognizing method named layout tree to cluster the similar layout visual blocks. The weight of all clusters is calculated, and the visual blocks in the cluster that has the highest weight are chosen as the data records to be extracted. The experiment results show that LTDE has higher effectiveness and better robustness for Web data extraction compared to previous works..
33. Hiroaki Ogata, Misato Oi, Kousuke Mohri, Fumiya Okubo, Atsushi Shimada, Masanori Yamada, Jingyun Wang, Sachio Hirokawa, Learning analytics for E-book-based educational big data in higher education, Smart Sensors at the IoT Frontier, 10.1007/978-3-319-55345-0_13, 327-350, 2017.05, [URL].
34. Naoya Onimura, Takanori Yamashita, Naoki Nakayama, Hidehisa Soejima, Sachio Hirokawa, Generation of sentence template graph from SOAP format medical documents, 2016 International Conference on Computational Science and Computational Intelligence, CSCI 2016
Proceedings - 2016 International Conference on Computational Science and Computational Intelligence, CSCI 2016
, 10.1109/CSCI.2016.0037, 159-162, 2017.03, [URL], Clinical pathways are a fixed list of viewpoints todescribe patients' condition. An outcome of a clinical pathwayconsists of several assessments of observation items andinspection items, all of which represented as numerical data. Onthe other hand, free description medical records in the SOAPformat are valuable data that indicate health care worker'sassessments. Standardization of textual description is a key issuewhen integrating these types of medical data. This study extractscharacteristic word 2-grams as source of sentence templates. Weapplied SVM (Support Vector Machine) using each outcomesas positive data. We constructed a search engine for 123736pain variance outcomes. We conducted empirical experimentsto confirm the usefulness of the proposed method..
35. Xibin Wang, Fengji Luo, Chunyan Sang, Jun Zeng, Sachio Hirokawa, Personalized movie recommendation system based on support vector machine and improved particle swarm optimization, IEICE Transactions on Information and Systems, 10.1587/transinf.2016EDP7054, E100D, 2, 285-293, 2017.02, [URL], With the rapid development of information andWeb technologies, people are facing 'information overload' in their daily lives. The personalized recommendation system (PRS) is an effective tool to assist users extract meaningful information from the big data. Collaborative filtering (CF) is one of the most widely used personalized recommendation techniques to recommend the personalized products for users. However, the conventional CF technique has some limitations, such as the low accuracy of of similarity calculation, cold start problem, etc. In this paper, a PRS model based on the Support Vector Machine (SVM) is proposed. The proposed model not only considers the items' content information, but also the users' demographic and behavior information to fully capture the users' interests and preferences. An improved Particle Swarm Optimization (PSO) algorithm is also proposed to improve the performance of the model. The efficiency of the proposed method is verified by multiple benchmark datasets..
36. Tetsuya Nakatoh, Kenta Nagatani, Toshiro Minami, Sachio Hirokawa, Takeshi Nanri, Miho Funamori, Analysis of the quality of academic papers by the words in abstracts, Thematic track on Human Interface and the Management of Information, held as part of the 19th International Conference on Human–Computer Interaction, HCI International 2017
Human Interface and the Management of Information
Supporting Learning, Decision-Making and Collaboration - 19th International Conference, HCI International 2017, Proceedings
, 10.1007/978-3-319-58524-6_34, 10274 LNCS, 434-443, 2017.01, [URL], The investigation of related research is very important for research activities. However, it is not easy to choose an appropriate and important academic paper from among the huge number of possible papers. The researcher searches by combining keywords and then selects an paper to be checked because it uses an index that can be evaluated. The citation count is commonly used as this index, but information about recently published papers cannot be obtained. This research attempted to identify good papers using only the words included in the abstract. We constructed a classifier by machine learning and evaluated it using cross validation. As a result, it was found that a certain degree of discrimination is possible..
37. Brendan Flanagan, Sachio Hirokawa, Emiko Kaneko, Emi Izumi, Classification of speaking proficiency level by machine learning and feature selection, 1st International Symposium on Emerging Technologies for Education, SETE 2016 Held in Conjunction with ICWL 2016
Emerging Technologies for Education - 1st International Symposium, SETE 2016 Held in Conjunction with ICWL 2016, Revised Selected Papers
, 10.1007/978-3-319-52836-6_72, 10108 LNCS, 677-682, 2017.01, [URL], Analysis of publicly available language learning corpora can be useful for extracting characteristic features of learners from different proficiency levels. This can then be used to support language learning research and the creation of educational resources. In this paper, we classify the words and parts of speech of transcripts from different speaking proficiency levels found in the NICT-JLE corpus. The characteristic features of learners who have the equivalent spoken proficiency of CEFR levels A1 through to B2 were extracted by analyzing the data with the support vector machine method. In particular, we apply feature selection to find a set of characteristic features that achieve optimal classification performance, which can be used to predict spoken learner proficiency..
38. Takanori Yamashita, Naoya Onimura, Hidehisa Soejima, Naoki Nakashima, Sachio Hirokawa, Graph clustering system for text-based records in a clinical pathway, 16th World Congress of Medical and Health Informatics: Precision Healthcare through Informatics, MedInfo 2017
MEDINFO 2017
Precision Healthcare through Informatics - Proceedings of the 16th World Congress on Medical and Health Informatics
, 10.3233/978-1-61499-830-3-649, 649-652, 2017.01, [URL], The progressive digitization of medical records has resulted in the accumulation of large amounts of data. Electronic medical data include structured numerical data and unstructured text data. Although text-based medical record processing has been researched, few studies contribute to medical practice. The analysis of unstructured text data can improve medical processes. Hence, this study presents a clustering approach for detecting typical patient's condition from text-based medical record of clinical pathway. In this approach, the sentences in a cluster are merged to generate a "sentence graph" of the cluster after classified feature word by Louvain method. An analysis of real text-based medical records indicates that sentence graphs can represent the medical treatment and patient's condition in a medical process. This method could help the standardization of text-based medical records and the recognition of feature medical processes for improving medical treatment..
39. Naoya Onimura, Takanori Yamashita, Hidehisa Soejima, Sachio Hirokawa, Generation of Sentence Template Graph from SOAP Format Medical Documents, Proc. CSCI2016, pp.159-162, 2016.11.
40. Takanori Yamashita, Yoshifumi Wakata, Hidehisa Soejima, Naoki Nakashima, Sachio Hirokawa, Structuralization of Variance Text Records in Clinical Pathway, Proc. APAMI2016, p.85, 2016.11.
41. Yusuke Adachi, Naoya Onimura, Takanori Yamashita, Sachio Hirokawa, Standard measure and SVM measure for feature selection and their performance effect for text classification, 18th International Conference on Information Integration and Web-Based Applications and Services, iiWAS 2016
18th International Conference on Information Integration and Web-Based Applications and Services, iiWAS 2016 - Proceedings
, 10.1145/3011141.3011190, Part F126325, 262-266, 2016.11, [URL], This paper compares the prediction performance of document classification based on a variety of feature selection measures. Empirical experiments were conducted for the dataset re0 with 10 measures for feature selection and with SVM. It is confirmed that the feature selection based on the SVM-score proposed by Sakai and Hirokawa (2012) outper-forms the standard measures with small number of features. In fact, 100 words are enough to get the similar performance obtained with all words. The reason of good performance of this feature selection is that the SVM-score capture not only the characteristic words of positive samples but of negative samples as well..
42. Wentao Li, Min Gao, Hua Li, Jun Zeng, Qingyu Xiong, Sachio Hirokawa, Shilling attack detection in recommender systems via selecting patterns analysis, IEICE Transactions on Information and Systems, 10.1587/transinf.2015EDP7500, E99D, 10, 2600-2611, 2016.10, [URL], Collaborative filtering (CF) has been widely used in recommender systems to generate personalized recommendations. However, recommender systems using CF are vulnerable to shilling attacks, in which attackers inject fake profiles to manipulate recommendation results. Thus, shilling attacks pose a threat to the credibility of recommender systems. Previous studies mainly derive features from characteristics of item ratings in user profiles to detect attackers, but the methods suffer from low accuracy when attackers adopt new rating patterns. To overcome this drawback, we derive features from properties of item popularity in user profiles, which are determined by users' different selecting patterns. This feature extraction method is based on the prior knowledge that attackers select items to rate with man-made rules while normal users do this according to their inner preferences. Then, machine learning classification approaches are exploited to make use of these features to detect and remove attackers. Experiment results on the MovieLens dataset and Amazon review dataset show that our proposed method improves detection performance. In addition, the results justify the practical value of features derived from selecting patterns..
43. Kiyota Hashimoto, Tasanawan Soonklang, Sachio Hirokawa, Feature Words of Moves in Scientific Abstracts, Proc. 5th IIAI-AAI,, pp.749-754, 2016.08.
44. Jun Zeng, Feng Li, Haiyang Liu, Junhao Wen, Sachio Hirokawa, A restaurant recommender system based on user preference and location in mobile environment, 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
Proceedings - 2016 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
, 10.1109/IIAI-AAI.2016.126, 55-60, 2016.08, [URL], Recommender system is an effective way to help users to obtain the personalized and useful information. However, due to complexity and dynamic, the traditional recommender system cannot work well in mobile environment. In this paper, we propose a restaurant recommender system in mobile environment. This recommender system adopts a user preference model by using the features of user's visited restaurants, and also utilizes the location information of user and restaurants to dynamically generate the recommendation results. Baidu map cloud service is used to implement the proposed recommender system. The result of a case study shows that the proposed restaurant recommender system can effectively utilize user's preference and the location information to recommend the personalized and suitable restaurants for different users..
45. Brendan Flanagan, Sachio Hirokawa, Automatic extraction and prediction of word order errors from language learning SNS, 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
Proceedings - 2016 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
, 10.1109/IIAI-AAI.2016.59, 292-295, 2016.08, [URL], Recent research into writing tools to support foreign language learners of English has focused on prevalent errors in learner writing, while other errors, such as word order errors, have received little attention. As the word order of some languages are similar, there are also large differences between languages which can affect foreign language learning. In this paper, we automatically extract corrected sentences that contain word order errors samples from a language learning SNS to create a word order error corpus for machine learning. This corpus is then analyzed to train and evaluate the effectiveness of SVM classifiers to automatically classify word order errors in learner writing..
46. Tetsuya Nakatoh, Hayato Nakanishi, Toshiro Minami, Kensuke Baba, Sachio Hirokawa, Bibliometric search with focused citation ratios, 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
Proceedings - 2016 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
, 10.1109/IIAI-AAI.2016.227, 150-153, 2016.08, [URL], A survey of related work is an important task for every researcher, and databases of scientific articles are indispensable for this task. This paper proposes a new visualization method for search results and demonstrates a system that implements this method. Given a query, the system returns a list of articles and displays a time series of citation counts (CCs) for each article. The novelty of the visualization is in its use of CC for the horizontal axis and focused CC (FCC) for the vertical axis. A scatter plot of the article reveals how the article was evaluated..
47. Kiyota Hashimoto, Tasanawan Soonklang, Sachio Hirokawa, Feature words of moves in scientific abstracts, 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
Proceedings - 2016 5th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2016
, 10.1109/IIAI-AAI.2016.38, 144-149, 2016.08, [URL], Extraction of structure from texts is a key issue of text mining. The rhetorical structure of move in scientific articles is useful for assisting in the reading and writing. In this paper, we classify move structure in the abstract of research articles with a small number of characteristic words that determine five moves of including background (B), purpose(P), method(M), result(R) and discussion(D). Eleven measures were introduced and used to select features of moves. Exhaustive parameter search were conducted to get the optimal combination of measure and the number of features. We applied support vector machine and evaluated 10 fold cross validations. The accuracies with optimal feature selections are 0.9022, 0.8322, 0.8442, 0.8820 and 0.8354 for B, P, M, R and D, respectively. They are 10% better than the baseline performance that use all keywords. This study surprisedly found that the negative feature words play central role for prediction performance improvement..
48. Tetsuya Nakatoh, Hayato Nakanishi, Toshiro Minami, Kensuke Baba, Sachio Hirokawa, A Visual Citation Search Engine,, Proc. HCI2016, 2016.07.
49. Brendan Franagan, Sachio Hirokawa, Support Vector Mind Map of Wine Speak, Proceedings of HCI2016, 2016.07.
50. Wentao Li, Min Gao, Hua Li, Jun Zeng, Qingyu Xiong, Sachio Hirokawa, Shilling Attack Detection in Recommender Systems via Selecting Patterns Analysis, 99-D(10), pp.2600-2611, 2016, 2016.07.
51. Brendan Franagan, Sachio Hirokawa, Automatic Extraction and Prediction of Word Order Errors From Language Learning SNS, Proc. 5th IIAI-AAI, pp. 292 - 295, 2016.07.
52. Jun Zeng, Feng Li, Haiyang Liu, Junhao Wen, Sachio Hirokawa, A Restaurant Recommender System Based on User Preference and Location in Mobile Environment, pp. 55 - 60, 2016.07.
53. Chao Zeng, Tetsuya Nakatoh, Hiroyuki Takeshita, Ryoji Hisadomi, Masanari Eguchi, Sachio Hirokawa, Discovery of Cultural Tourism Preference in Multilingual Tourism Information Site, Proc. ACIS2016, pp.129-134, 2016.07.
54. Jie Zou, Ling Xu, Mengning Yang, Xiaohong Zhang, Jun Zeng, Sachio Hirokawa, Automated duplicate bug report detection using multi-factor analysis, IEICE Transactions on Information and Systems, 10.1587/transinf.2016EDP7052, E99D, 7, 1762-1775, 2016.07, [URL], The bug reports expressed in natural language text usually suffer from vast, ambiguous and poorly written, which causes the challenge to the duplicate bug reports detection. Current automatic duplicate bug reports detection techniques have mainly focused on textual information and ignored some useful factors. To improve the detection accuracy, in this paper, we propose a new approach calls LNG (LDA and N-gram) model which takes advantages of the topic model LDA and word-based model Ngram. The LNG considers multiple factors, including textual information, semantic correlation, word order, contextual connections, and categorial information, that potentially affect the detection accuracy. Besides, the Ngram adopted in our LNG model is improved by modifying the similarity algorithm. The experiment is conducted under more than 230,000 real bug reports of the Eclipse project. In the evaluation, we propose a new evaluationmetric, namely exact-accuracy (EA) rate, which can be used to enhance the understanding of the performance of duplicates detection. The evaluation results show that all the recall rate, precision rate, and EA rate of the proposed method are higher than treating them separately. Also, the recall rate is improved by 2.96%-10.53% compared to the state-of-art approach DBTM..
55. Fumiya Okubo, Sachio Hirokawa, Miaato Oi, CHENGJIU YIN, Atsushi Shimada, Kojima Kentaro, Masanori Yamada, Hiroaki Ogata, Learning Activity Features of High Performance Students, Proc. LAK2016, 2016.04.
56. Sachio Hirokawa, Tetsuya Nakatoh, Hayato Nakanishi, Accumulated citation count as fertileness of scientific article, International Conference on Computational Science and Computational Intelligence, CSCI 2015
Proceedings - 2015 International Conference on Computational Science and Computational Intelligence, CSCI 2015
, 10.1109/CSCI.2015.74, 119-122, 2016.03, [URL], The literature survey by scientific bibliographic data base is indispensable in the research activities. We can find related articles with appropriate keywords. However, the threads of related research are not easy to grasp from the search result. It is necessary to repeat a search, judge a citation relation and figure out the thread. The present paper proposes the index "accumulated citation count" of a scientific article to measure the thread of citations that starts from the article..
57. Takanori Yamashita, Yoshifumi Wakata, Naoki Nakashima, Sachio Hirokawa, Performance Evaluation of Predicting Period of Hospitalization from Operation Record, Proc. BHI2016 (International Conference on Biomedical and Health Informatics), 2016.02, 医療現場で情報デジタル化により、医療情報の二次利用が期待されている。特に、医師や看護師による自由自由記述文書は高度な判断が含まれ貴重なデータであるが、分析は容易でない。本論文では、機械学習を適用することで、人工股間節置換術の記録から、対象患者の長期入院の可能性を高い精度で推定することに成功した。また、長期入院となる要因をクリニカルパスの構造に着目して明らかにした。.
58. Takuya Hirao, Takahiko Suzuki, Sachio Hirokawa, Nao Wariishi, Kyota Hashimoto, Evaluation of Integrity of WordNet by Combining Word Similarity and Random Forest, Proc. AROB2016, pp.739-743, 2016.01.
59. Hayato Nakanishi, Tetsuya Nakatoh, Sachio Hirokawa, Steep Increase Trigger of Citation, Proc. AROB2016, pp.744-748, 2016.01.
60. Brendan Flanagan, Sachio Hirokawa, Carita Paradis, Kiyota Hashimoto, Analysis of the diachronic relations of adjective antonym pairs in wine tasting notes, Proc. AROB2016, pp.749-754, 2016.01.
61. Hiroto Nakae, Hitoshi INOUE, Kazuhisa Noguchi, Kiyota Hashimoto, Akira Aaiba, Sachio Hirokawa, Generation and Evaluation of Quizzes from Manyo-Shu and Kokin-Waka-Shu, Proc. AROB2016, pp.755-758, 2016.01.
62. Tetsuya Nakatoh, Kiyota Hashimoto, Sachio Hirokawa, Analysis of Infrequent Words in Tourism Blogs, Proc. AROB2016, pp.597-601, 2016.01.
63. Sachiko Nakajima, Yukiko Watanabe, Sachio Hirokawa, Survey on Japanese Academic Library Reference Services, Proc. AROB2016, pp.403-408, 2016.01.
64. Yuusuke Adachi, Takanori Yamashita, Yoshifumi Wakata, Hidehisa Soejima, Yosifumi Wakata, Sachio Hirokawa, Comparison of SVM and Decision Tree for Prediction of Postoperative Hospital Stay, Proc. AROB2016, pp.414-419, 2016.01.
65. Naoya Onimura, Brendan Flanagan, Takanori Yamashita, Sachio Hirokawa, Performance Effect of Feature Selection on Support Vector Machine, Proc. AROB2016, pp.420-424, 2016.01.
66. Yuusuke Yoshida, Takahiko Suzuki, Kyota Hashimoto, Sachio Hirokawa, Correspondence of Clustering of Questions and Clustering of Answers, Proc. AROB2016, pp.425-429, 2016.01.
67. Nao Wariishi, Shuichi Mitarai, Takahiko Suzuki, Sachio Hirokawa, Text Mining of Daily Sales Reports,, Proc. AROB2016, pp.430-435, 2016.01.
68. Tetsuya Nakatoh, Hayato Nakanishi, Toshiro Minami, Kensuke Baba, Sachio Hirokawa, A visual citation search engine, 18th International Conference on Human-Computer Interaction, HCI International 2016
Human Interface and the Management of Information
Information, Design and Interaction - 18th International Conference, HCI International 2016, Proceedings
, 10.1007/978-3-319-40349-6_17, 9734, 168-178, 2016.01, [URL], Carrying out the survey of the related researches is an essential part in research activities, the aim of which is to have an overall view of the target field. Generally, we take two approaches toward this aim. One approach is paying attention to selected articles and deeply investigate them. The selection is performed according to some indicators for measuring importance. The other approach is considering the citation relations between articles. One problem is that these approaches cannot be combined straightforwardly. Another problem in carrying out the survey is that there are a huge amount of articles exist already. The aim of this paper is to propose a framework of a visualization system that assists us in surveying related researches. The system displays the important articles together with their key citation relations by displaying not only direct citations between important articles but also the indirect, or weak-tie, citation relations that connect them..
69. Brendan Flanagan, Sachio Hirokawa, Correlation between an Entropy Based Measure and English Language Learner Proficiency, 4th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2015
Proceedings - 2015 IIAI 4th International Congress on Advanced Applied Informatics, IIAI-AAI 2015
, 10.1109/IIAI-AAI.2015.288, 349-353, 2016.01, [URL], It is important for education systems to analyze and provide an appropriate level of support to meet the needs of learners. An example of this is how the effectiveness of automatic language learner error detection and correction can vary depending on the learner's proficiency level. Covering a wide range of language complexity makes the task of error detection difficult. By predicting the learner's proficiency level, different error models can be applied for different proficiency levels. In this paper, we propose a measure based on the frequency of words in the sentences produced by learners during speaking exams to predict the learner's language proficiency. The proposed measure is compared to the learner's vocabulary size by correlation analysis. The results suggest that there is a stronger correlation between the proposed measure and the proficiency of the learner than the learner's vocabulary size..
70. Tetsuya Nakatoh, Hayato Nakanishi, Kensuke Baba, Sachio Hirokawa, Focused Citation Count
A Combined Measure of Relevancy and Quality, 4th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2015
Proceedings - 2015 IIAI 4th International Congress on Advanced Applied Informatics, IIAI-AAI 2015
, 10.1109/IIAI-AAI.2015.282, 166-170, 2016.01, [URL], Literature survey of scientific articles depends on the relevancy and the quality of the obtained list. Relevancy might be controlled by an appropriate search query and the relevancy ranking of the search result. Citation count (CC) is widely used and useful as an easy measure to evaluate the quality of articles. However, articles with high citation count might cover a wide area, while they might have the low relation to a query. Moreover, relevancy and citation count are two independent measures that we cannot choose at the same time. The present paper proposes 'Focused Citation Count(FCC)', a novel measure that focuses only on the relevant articles to count the citation. We realize the integration of relevancy and quality by restricting the articles that cite the target article. Empirical evaluation was conducted with 10,186 articles on 'bibliometrics' by P@N measure, the average precision at top N search result. It is confirmed that the ranking by the proposed method FCC gained over 0.8 and outperformed the conventional ranking by CC whose score was below 0.6..
71. Fumiya Okubo, Sachio Hirokawa, Misato Terai, Atsushi Shimada, Kojima Kentaro, Masanori Yamada, Hiroaki Ogata, Learning activity features of high performance students, 1st International Workshop on Learning Analytics Across Physical and Digital Spaces, CrossLAK 2016
CEUR Workshop Proceedings
, 1601, 28-33, 2016.01, In this paper, we present a method of identifying learning activities that are important for students to achieve good grades. For this purpose, the data of 99 students were collected from a learning management system and an e-book system, including attendance, time on preparation and review, submission of reports, and quiz scores. We applied a support vector machine to these data to calculate a score of importance for each learning activity reflecting its contribution to the attainment of an A grade. Selecting certain important learning activities by following several evaluation measures, we verified that these learning activities played a crucial role in predicting final student achievements. One of the obtained results implies that time on preparation and review in the middle part of a course influences a student's final achievement..
72. Chengjiu Yin, Jane Yin Kim Yau, Noriko Uosaki, Sachio Hirokawa, Etsuko Kumamoto, Measuring & evaluating digital textbooks through quizzes, 24th International Conference on Computers in Education, ICCE 2016
ICCE 2016 - 24th International Conference on Computers in Education
Think Global Act Local - Main Conference Proceedings
, 374-379, 2016.01, We currently utilize the Moodle learning management system for teachers and students who participate in the course 'College of Liberal Arts and Sciences' at Kobe University in Japan. Digital textbooks, reports, quizzes and questionnaires in this course were administered using Moodle. In this paper, we proposed to use quizzes to measure and evaluate those digital textbooks recorded on Moodle. At the beginning of our study, we examined the questions that students got lower scores, and then we found the related teaching materials of digital textbooks and feedback to the teachers in order to improve the content of these digital textbooks..
73. Sachio Hirokawa, Message from Congress General Chair, Quaternary International, 10.1109/IIAI-AAI.2015.152, xvii, 2016.01, [URL].
74. Yanting Xu, Sachio Hirokawa, Hitoshi Inoue, Procurement Service of Japanese Product between Japan and China, 1st International Conference on Computer Application Technologies, CCATS 2015
Proceedings - 2015 International Conference on Computer Application Technologies, CCATS 2015
, 10.1109/CCATS.2015.18, 38-39, 2016.01, [URL], In this paper, firstly we describe difference of information provision and management in Japan-China Internet business. Next we state the social problem in China. Finally we summarize procurement service of Japanese products between Japan and China..
75. Nao Wariishi, Brendan Flanagan, Takahiko Suzuki, Sachio Hirokawa, Sentiment Analysis of Wine Aroma, 4th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2015
Proceedings - 2015 IIAI 4th International Congress on Advanced Applied Informatics, IIAI-AAI 2015
, 10.1109/IIAI-AAI.2015.253, 207-212, 2016.01, [URL], It has been easy for us to send information thanks to the growth of the Internet. Information includes reputation. Analysis of the large amount of reputation and making summary of the analysis is very helpful for consumers to decide which goods they should buy. They are also helpful for companies in order to make marketing decisions. Many researches try to classify documents on whether they have positive or negative sentiment. In this paper, we focus on more complex sentiment. We will show a result of machine classification of wine reviews from the point of view of 'aroma' using Support Vector Machine (SVM)..
76. Brendan Flanagan, Sachio Hirokawa, Support vector mind map of wine speak, 18th International Conference on Human-Computer Interaction, HCI International 2016
Human Interface and the Management of Information
Information, Design and Interaction - 18th International Conference, HCI International 2016, Proceedings
, 10.1007/978-3-319-40349-6_13, 127-135, 2016.01, [URL], Models created by blackbox machine learning techniques such as SVM can be difficult to interpret. It is because these methods do not offer a clear explanation of how classifications are derived that is easy for humans to understand. Other machine learning techniques, such as: decision trees, produce models that are intuitive for humans to interpret. However, there are often cases where an SVM model will out preform a more intuitive model, making interpretation of SVM trained models an important problem. In this paper, we propose a method of visualizing linear SVM models for text classification by analyzing the relation of features in the support vectors. An example of this method is shown in a case study into the interpretation of a model trained on wine tasting notes..
77. Yasuhiro Yamada, Daisuke Ikeda, Sachio Hirokawa, Unique Links as Weak Ties, 4th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2015
Proceedings - 2015 IIAI 4th International Congress on Advanced Applied Informatics, IIAI-AAI 2015
, 10.1109/IIAI-AAI.2015.266, 132-136, 2016.01, [URL], It is important to find suitable partners in order to form successful collaborations between companies and university researchers. We consider finding the partners by calculating the similarity of the documents such as scientific papers and patents. We focus on weak (unique) links of researchers as the local similarity of their documents, instead of strong links as the global similarity of the documents. In the present paper, we propose a system that matches partners using documents such as research papers and patents. Given a query, the proposed system outputs a graph of unique research in retrieved documents. Each node in the graph corresponds to a word with a document frequency of two. Two words connected by an edge occur in the same two documents, and neither word appears in other retrieved documents. The edge is labeled with the names of the researchers involved in the documents in which the two words appear. Experiments are conducted using graphs output by the system..
78. Takuya Hirao, Nao Wariishi, Takahiko Suzuki, Sachio Hirokawa, Vector Similarity of Related Words in the Japanese Word Net, 4th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2015
Proceedings - 2015 IIAI 4th International Congress on Advanced Applied Informatics, IIAI-AAI 2015
, 10.1109/IIAI-AAI.2015.254, 142-147, 2016.01, [URL], Word2vec is a tool that produces vector representation of words from a large amount of text data. In this paper, we show that only a part of the vector space produced by word2vec is enough to represent the collective sense of a set of related words in the Japanese Word Net. Further, we will show that there is a subspace in the vector space which do not relate to the collective sense. We construct a compact decision tree by using the vectors in order to distinguish whether a given word belongs to the set of related words..
79. Satoshi V. Suzuki, Sachio Hirokawa, Shinji Mukoyama, Ryo Uehara, Hiroyuki Ogata, Student behavior in computer simulation practices by pair programming and flip teaching, 24th International Conference on Computers in Education, ICCE 2016
ICCE 2016 - 24th International Conference on Computers in Education
Think Global Act Local - Main Conference Proceedings
, 212-221, 2016, Recent education roles encourage willingness to learn individually, solve unfamiliar problems using knowledge acquired through information and communication technologies (ICTs), and collaborate with others. Various learning methods to cultivate this ability have been invented and researchers discussed learning effect on such methods with qualitative and quantitative analyses. One of the authors introduced pair programming as a method of peer learning and flip teaching, which consists of preliminary learning of basic programming and advanced learning practices based on peer activity in classroom lessons, into computer simulation practices for undergraduate students. With introducing class schedule design for flip teaching and development of peer learning preparation support system for determining the appropriate pair formation and seat allocation in the classroom utilizing a probabilistic combinatorial optimization algorithm, this study focuses on learning behavior of wellperforming students, analyzing learning records and access logs on a learning management system and answers to a questionnaire administered after the practices. In this analysis, we attempted to discriminate behavior of well-performing students observed from the learning records, access logs, and questionnaire to discover best practices for improving the performance of medium to bottom-line students. Well-performing students tended to prepare for classroom lessons in good time, but their performance depended on the lesson content difficulty. A correlation was observed between the frequency of interaction among students and skill acquisition. We discuss how to improve learning environments using ICTs and collaborative learning methods based on the analysis..
80. Brendan Flanagan, Kiyota Hashimoto, Sachio Hirokawa, Analysis of Antonymic Adjective Meaning Dimensions in Winespeak, Proc. SNLP2016, CD, 2015.12.
81. Toshiro Minami, Sachio Hirokawa, Yoko Ohura, Kiyota Hashimoto, Influence Analysis of Parts of Speech to Examination Score using Texts from Students' Self/Lecture Evaluation, Proc. SNLP2016, CD, 2015.12.
82. CHENGJIU YIN, Fumiya Okubo, Miaato Oi, Sachio Hirokawa, Masanori Yamada, Kojima Kentaro, Hiroaki Ogata, Analyzing the Features of Learning Behaviors of Students using e-Books, Proc. ICCE2015, pp.617-626, 2015.12.
83. Sachio Hirokawa, CHENGJIU YIN, JINGYUN WANG, Misato OI, Hiroaki Ogata, Visualization of e-Book Learning Logs, Proc. ICCE2015, pp.659-664, 2015.12.
84. CHENGJIU YIN, Fumiya Okubo, Atsushi Shimada, Sachio Hirokawa, Hiroaki Ogata, Misato OI, Identifying and Analyzing the Learning Behaviors of Students using e‐Books, Proc. ICCE2015, pp.118-120, 2015.12.
85. Sachio Hirokawa, Tetsuya Nakatoh, Hayato Nakanishi, Accumulated Citation Count as Fertileness of Scientific Article, 119-122, 2015.12, Literature survey of scientific articles depends on the relevancy and
the quality of the obtained list. Relevancy might be controlled by an
appropriate search query and the relevancy ranking of the search
result. Citation count (CC) is widely used and useful as an easy
measure to evaluate the quality of articles. However, articles with
high a citation score might cover a wide area or have low relevancy to
the query. Moreover, relevancy and citation count are two independent
measures that we cannot chose at the same time.
The present paper proposes "Focused Citation Count, a novel measure
(FCC)" that focuses only on the relevant articles to count the
citation. We realize the integration of relevancy and quality by
restricting the articles that cite the target article. Empirical
evaluation was conducted with 10,186 articles on ``bibliometrics'' by
P@N measure, the average precision at top N search result. It is
confirmed that the ranking by the proposed method FCC gained over 0.8
and outperformed the conventional ranking by CC whose score was below
0.6..
86. Tetsuya Nakatoh, Hayato Nakanishi, Toshiro Minami, Sachio Hirokawa, Threads and History of Bibliometrics, 27-31, 2015.12.
87. Hayato Nakanishi, Tetsuya Nakatoh, Sachio Hirokawa, Cause Analysis for Steep Increase of Citation, Proc. KICSS2015, 2015.11, The literature survey is indispensable in scientific activities. We have to
choose efficiently new articles and good articles from the articles obtained
by search engine. However, we cannot read or comprehend the contents of all
articles due to the explosion of the number of articles being published. The
citation count of an article is convenient as quantitative measure for
evaluation of the article. The present paper is an experiment of a hypothesis
that the evaluation of an article is determined with respect to the evaluation
of articles that cite the article. The authors confirms the hypothesis by time
series analysis of citation counts. Three articles with steep increase were
chosen as experiments in the field of ``bibliometrics.''.
88. Haruka Kubo, Takanori Yamashita, Brendan Flanagan, Yoshifumi Wakata, Naoki Nakashim, Sachio Hirokawa, Feature Words to Predict Long Post-Operatively Stay in Semi-structured Medical Records, Proceedings of ACIS2015, 2015.10, The number of medical records of POMR (Problem Oriented Medical Record) format is increasing for the quantitative evaluation and improvement of medical process. A POMR format record consists of patients' subjective observation (S), objective items of test results (O), assessment, evaluation and judgement (A), future's treatment policy (P) and free de-scription (F). The present paper applied SVM(support vector machine) to extract characteristic words to predict the patients of long post-operatively stay. Analysis of 3,840 medical records of Saiseikai Kumamoto Hos-pital revealed that the objective component contains the most crucial words..
89. Haruka Kubo, Takanori Yamashita, Yoshifumi Wakata, Naoki Nakashim, Sachio Hirokawa, Effect of Synonym on Prediction Performance for Postoperative Hospital Stay by Text Mining, Proc. AROB2016, pp.409-413, 2015.10.
90. Takanori Yamashita, Yoshifumi Wakata, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Brendan Flanagan, Naoki Nakashim, Sachio Hirokawa, Visualization of Key factor Relation in Clinical Pathway, Proceedings of KES2015, 342-351, 2015.09.
91. 合田 和正, 峯 恒憲, 廣川 佐千男, 学習態度に関する自己評価記述の正確さと成績推定性能の相関, 電子情報通信学会論文誌 D, J98-D, 9, 192-202, 2015.09.
92. 廣川 佐千男, 伊東栄典, 馬場謙介, 関連研究探索のための検索可視化システム, 情報管理, 58, 6, 447-454, 2015.09, 科学技術の加速度的発達により、一般社会と専門家の乖離は大きく、若い人の理系離れも
問題となっている。専門家であっても、複合領域や未知の分野の調査は容易ではない。本
稿では、我が国の科学技術の基本情報である科学研究費の採択課題概要を対象とした検索
可視化システムを紹介する。本システムでは、概要に現れる単語だけでなく、専門用語、
領域分類、研究者名、研究機関名、年度などの単語を異なる色の関連マップとして表示す
る。単語の属性識別により関連解釈が可能となり、知りたいテーマに関連して、「だれ
が、どこで、どんな」研究活動を行っているかを把握できる。本稿ではシステムの概要
と、探索的調査の事例を紹介する。.
93. Brendan Franagan, Sachio Hirokawa, Correlation Between an Entropy Based Measure and English Language Learner Proficiency, Proceedings of HCI2015, 2015.08, It has been easy for us to send information thanks to the growth of
the Internet. Information includes reputation. Analysis of the large
amount of reputation and making summary of the analysis is very
helpful for consumers to decide which goods they should buy. They are
also helpful for companies in order to make marketing decisions. Many
researches try to classify documents whether they have positive or
negative sentiment. In this paper, we focus on more complex
sentiment. We will show a result of machine classification of wine
reviews on a point of view of "aroma"using Support Vector Machine
(SVM)..
94. Brendan Franagan, Nao Wariishi, Takahiko Suzuki, Sachio Hirokawa, Predicting and Visualizing Wine Characteristics Through Analysis of Tasting Notes From Viewpoints, Proceedings of HCI2015, 2015.08, When describing complex characteristics of a specific genre,
specialist expressions are often used. This can become quite a
problematic situation for an inexperienced person, as expressions not
used i n everyday language are difficult to understand. This is
particularly apparent when trying to describe wines, known as
winespeak, as a range of specialist expressions are used in a
subjective manner. In this paper, we propose that the descriptions of
wines can be analyzed from various points of view to automatically
predict and visualize the sensory sentiment characteristics described
within the expressions as a radar chart. This would enable those not
knowledgeable in winespeak to visualize and co m pare the complex
descriptions often found in expert tasting notes.
95. Yanting XU, Hitoshi INOUE, Sachio Hirokawa, Procurement Service of Japanese Product between Japan and China, Proc. ICCAT2015, 2015.08.
96. Takanori Yamashita, Yoshifumi Wakata, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Brendan Flanagan, Naoki Nakashim, Sachio Hirokawa, Temporal Relation Extraction in Outcome Variance of Clinical Pathway, Proceedings of MEDINFO2015, 2015.08, Recently, the clinical pathway has been progressed with
digi-talization and the analysis of activity. There are many previ-ous
researches on the clinical pathway, however not many directly feedback
to medical practice. We constructed a Mind map system that applies the
spanning tree. This system could visualize temporal relations in the
outcome variances, and extracted outcomes affect long-term
hospitalization..
97. Brendan Franagan, Sachio Hirokawa, Web of Wine Words: Hierarchy Visualization of Wine Speak by Restricted Bootstrap, Proceedings of DPTA15, 290-296, 2015.08, Visualization of the relation of characteristic words can be useful
for interpreting search results and enable comparisons to be made
between mu ltiple searches. In this paper we introduce a method of
analysis by applying restricted bootstrapping to a set of
characteristic words for extracting specificity or generality
relations. These relations are used to construct a tree structure of
the charact eristic words that represents their hierarchical
specificity or generality. This method was applied to a corpus of wine
tasting notes to identify the characteristics of two wine regions by
hierarchy tree. The results are compared with the frequencies of the
characteristic words..
98. Tetsuya Nakatoh, Sachio Hirokawa, Extraction of Tourism Objects from Blogs, 10.1007/978-3-662-47227-9_4., 2015.08.
99. Hayato Nakanishi, Tetsuya Nakatoh, Kensuke Baba, Sachio Hirokawa, Focused Citation Count -- A Combined Measure of Relevancy and Quality, Proc. AAI2015, 2015.07, Literature survey of scientific articles depends on the relevancy and
the quality of the obtained list. Relevancy might be controlled by an
appropriate search query and the relevancy ranking of the search
result. Citation count (CC) is widely used and useful as an easy
measure to evaluate the quality of articles. However, articles with
high a citation score might cover a wide area or have low relevancy to
the query. Moreover, relevancy and citation count are two independent
measures that we cannot chose at the same time.
The present paper proposes "Focused Citation Count, a novel measure
(FCC)" that focuses only on the relevant articles to count the
citation. We realize the integration of relevancy and quality by
restricting the articles that cite the target article. Empirical
evaluation was conducted with 10,186 articles on ``bibliometrics'' by
P@N measure, the average precision at top N search result. It is
confirmed that the ranking by the proposed method FCC gained over 0.8
and outperformed the conventional ranking by CC whose score was below
0.6..
100. Tetsuya Nakatoh, Hayato Nakanishi, Sachio Hirokawa, Journal Impact Factor Revised with Focused View, Proc. 7th KES International Conference on Intelligent Decision Technologies (KES-IDT 2015), 471-481, 2015.07, The evaluation of an article is an important issue for scientific
research. A journal impact factor is used widely to evaluate the
impact of the journals. An impact factor is considered as an influence
measure of articles. Articles in a high impact factor journal tend to
have strong impact on the wide research fields. However, this does not
imply that they have big influence in a particular research
area. Measuring the speciality and the generality of research results
is not trivial task. The present paper proposes a generalized method
to evaluate the influ- ence of a journal with respect to a focused
view. An empirical evaluation was conducted on "bibliometrics" related
10,186 articles..
101. Nao Wariishi, Brendan Franagan, Takahiko Suzuki, Sachio Hirokawa, Sentiment Analysis of Wine Aroma, Proceedings of AAI2015, 207-214, 2015.07, It has been easy for us to send information thanks to the growth of the
Internet. Information includes reputation. Analysis of the large amount of
reputation and making summary of the analysis is very helpful for consumers to
decide which goods they should buy. They are also helpful for companies in order
to make marketing decisions. Many researches try to classify documents
whether~they have positive or negative sentiment. In this paper, we focus on
more complex sentiment. We will show a result of machine classification of wine
reviews on a point of view of “aroma” using Support Vector Machine (SVM).
Keywords―sentiment analysis, SVM, feature selection.
102. Yasuhiro Yamada, Daisuke Ikeda, Sachio Hirokawa, Unique Links as Weak Ties, Proc. AAI2015, 2015.07.
103. Takuya Hirao, Takahiko Suzuki, Sachio Hirokawa, Vector Similarity of Related Words in the Japanese WordNet, Proc. AAI2015, 31-36, 2015.07, Word2vec is a tool that produces vector representation of words from a
large amount of text data. In this paper, we show that only a part
of the vector space produced by word2vec is en ough to represent the
collective sense of a set of related word s in the Japanese WordNet
. Further, we will show that there is a subspace in the vector space
which do not relate to the collective sense. We construct a compact
decision tree by using the vect ors in order to distinguish whether a
given word belongs to the set of related words..
104. Shaymaa Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, A Predictive Model to Evaluate Students Performance, Journal of Information Processing, 23, 2, 192-202, 2015.02.
105. Zhuo Jiang, Junhao Wen, Jun Zeng, Yihao Zhang, Xibin Wang, Sachio Hirokawa, Dynamic macro-based heuristic planning through action relationship analysis, IEICE Transactions on Information and Systems, 10.1587/transinf.2014EDP7170, E98D, 2, 363-371, 2015.02, [URL], The success of heuristic search in AI planning largely depends on the design of the heuristic. On the other hand, previous experience contains potential domain information that can assist the planning process. In this context, we have studied dynamic macro-based heuristic planning through action relationship analysis. We present an approach for analyzing the action relationship and design an algorithm that learns macros in solved cases. We then propose a dynamic macro-based heuristic that appropriately reuses the macros rather than immediately assigning them to domains. The above ideas are incorporated into a working planning system called Dynamic Macro-based Fast Forward planner. Finally, we evaluate our method in a series of experiments. Our method effectively optimizes planning since it reduces the result length by an average of 10% relative to the FF, in a time-economic manner. The efficiency is especially improved when invoking an action consumes time..
106. Shaymaa E. Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, Predicting students' grades based on free style comments data by artificial neural network, 44th Annual Frontiers in Education Conference, FIE 2014
Proceedings - Frontiers in Education Conference, FIE
, 10.1109/FIE.2014.7044399, 2015-February, February, 2015.02, [URL], Predicting students' academic achievement with high accuracy has an important vital role in many academic disciplines. Most recent studies indicate the important role of the data type selection. They also attempt to understand individual students more deeply by analyzing questionnaire for a particular purpose. The present study uses free-style comments written by students after each lesson, to predict their performance. These comments reflect their learning attitudes to the lesson, understanding of subjects, difficulties to learn, and learning activities in the classroom. To reveal the high accuracy of predicting student's grade, we employ (LSA) latent semantic analysis technique to extract semantic information from students' comments by using statistically derived conceptual indices instead of individual words, then apply (ANN) artificial neural network model to the analyzed comments for predicting students' performance. We chose five grades instead of the mark itself to predict student's final result. Our proposed method averagely achieves 82.6% and 76.1% prediction accuracy and F-measure of students' grades, respectively..
107. Shaymaa E. Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, A predictive model to evaluate student performance, Journal of information processing, 10.2197/ipsjjip.23.192, 23, 2, 192-201, 2015.01, [URL], In this paper we propose a new approach based on text mining techniques for predicting student performance using LSA (latent semantic analysis) and K-means clustering methods. The present study uses free-style comments written by students after each lesson. Since the potentials of these comments can reflect student learning attitudes, understanding of subjects and difficulties of the lessons, they enable teachers to grasp the tendencies of student learning activities. To improve our basic approach using LSA and k-means, overlap and similarity measuring methods are proposed. We conducted experiments to validate our proposed methods. The experimental results reported a model of student academic performance predictors by analyzing their comments data as variables of predictors. Our proposed methods achieved an average 66.4% prediction accuracy after applying the k-means clustering method and those were 73.6% and 78.5% by adding the overlap method and the similarity measuring method, respectively..
108. Chengjiu Yin, Fumiya Okubo, Atsushi Shimada, Misato Terai, Sachio Hirokawa, Masanori Yamada, Kojima Kentaro, Hiroaki Ogata, Analyzing the features of learning behaviors of students using e-Books, 23rd International Conference on Computers in Education, ICCE 2015
Workshop Proceedings of the 23rd International Conference on Computers in Education, ICCE 2015
, 616-626, 2015.01, The analysis of learning behavior and identification of learning style from learning logs are expected to benefit instructors and learners. This study describes methods for processing learning logs, such as data collection, integration, and cleansing, developed in Kyushu University. The research aims to analyze learning behavior and identify students' learning style using student's learning logs. Students were clustered into four groups using k-means clustering, and features of their learning behavior were analyzed in detail. We found that Digital Backtrack Learning style is better than Digital Sequential Learning style..
109. Chengjiu Yin, Fumiya Okubo, Atsushi Shimada, Misato Terai, Sachio Hirokawa, Hiroaki Ogata, Identifying and analyzing the learning behaviors of students using e-books, 23rd International Conference on Computers in Education, ICCE 2015
Proceedings of the 23rd International Conference on Computers in Education, ICCE 2015
, 118-120, 2015.01, Analyses on students' learning behaviors comprise an important thrust in education research. This study focused on e-books system used in the classroom and this system recorded students' learning logs in their daily academic life. These learning logs can be used to analysis students' learning behaviors. By performing partial correlation analysis, the study found that a number of learning behaviors have a significant relation with students' test scores..
110. Tetsuya Nakatoh, Hayato Nakanishi, Sachio Hirokawa, Journal impact factor revised with focused view, 7th KES International Conference on Intelligent Decision Technologies, KES-IDT 2015
Intelligent Decision Technologies - Proceedings of the 7th KES International Conference on Intelligent Decision Technologies, KES-IDT 2015
, 10.1007/978-3-319-19857-6_40, 39, 471-481, 2015.01, [URL], The evaluation of an article is an important issue for scientific research. A journal impact factor is used widely to evaluate the impact of the journals. An impact factor is considered as an influence measure of articles. Articles in a high impact factor journal tend to have strong impact on the wide research fields. However, this does not imply that they have big influence in a particular research area. Measuring the speciality and the generality of research results is not trivial task. The present paper proposes a generalized method to evaluate the influence of a journal with respect to a focused view. An empirical evaluation was conducted on “bibliometrics” related 10,186 articles..
111. Brendan Flanagan, Nao Wariishi, Takahiko Suzuki, Sachio Hirokawa, Predicting and visualizing wine characteristics through analysis of tasting notes from viewpoints, 17th International Conference on Human Computer Interaction, HCI 2015
HCI International 2015 – Posters Extended Abstracts - International Conference, HCI International 2015, Proceedings
, 10.1007/978-3-319-21380-4_104, 528, 613-619, 2015.01, [URL], When describing complex characteristics of a specific genre, specialist expressions are often used. This can become quite a problematic situation for an inexperienced person, as expressions not used in everyday language are difficult to understand. This is particularly apparent when trying to describe wines, known as winespeak, as a range of specialist expressions are used in a subjective manner. In this paper, we propose that the descriptions of wines can be analyzed from various points of view to automatically predict and visualize the sensory sentiment characteristics described within the expressions as a radar chart. This would enable those not knowledgeable in winespeak to visualize and compare the complex descriptions often found in expert tasting notes..
112. Brendan Flanagan, Chengjiu Yin, Takahiko Suzuki, Sachio Hirokawa, Prediction of learner native language by writing error pattern, 2nd International Conference on Learning and Collaboration Technologies, LCT 2015 Held as Part of 17th International Conference on Human-Computer Interaction, HCI International 2015
Learning and Collaboration Technologies - 2nd International Conference, LCT 2015 Held as Part of HCI International 2015, Proceedings
, 10.1007/978-3-319-20609-7_9, 9192, 87-96, 2015.01, [URL], The native language of a foreign language learner can have an effect on the errors they make because of similarities or differences between the two languages. In order to provide effective error prediction and correction for nonnative English language learners it is important to identify their specific characteristic error patterns that are influenced by their native language. In this paper, we examine analyzing error detection scores to predict the native language of an English language learner. 15 categories of error detection scores are combined to create an error prediction score vector representation of each sentence. The native language is predicted by training an SVM classifier with the error vectors. The results are compared to an SVM classifier trained with just word representations of the learner writing sentences..
113. Takanori Yamashita, Yoshifumi Wakata, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Brendan Franagan, Naoki Nakashima, Sachio Hirokawa, Temporal Relation Extraction in Outcome Variances of Clinical Pathways, 15th World Congress on Health and Biomedical Informatics, MEDINFO 2015
MEDINFO 2015
eHealth-Enabled Health - Proceedings of the 15th World Congress on Health and Biomedical Informatics
, 10.3233/978-1-61499-564-7-1077, 2015.01, [URL], Recently the clinical pathway has progressed with digitalization and the analysis of activity. There are many previous studies on the clinical pathway but not many feed directly into medical practice. We constructed a mind map system that applies the spanning tree. This system can visualize temporal relations in outcome variances, and indicate outcomes that affect long-term hospitalization..
114. Sachio Hirokawa, Chengjiu Yin, Jingyun Wang, Misato Terai, Hiroaki Ogata, Visualization of e-Book learning logs, 23rd International Conference on Computers in Education, ICCE 2015
Workshop Proceedings of the 23rd International Conference on Computers in Education, ICCE 2015
, 659-664, 2015.01, Learning environment with e-book enables learners to learn anytime and anywhere they like according to their own pace. There is a large expectation on e-book as personal learning tool. Understanding and grasping the learning status of students is crucial matter for teachers and for the learning system. Access log of e-books should be basis for analyzing the learning behavior. The authors are constructing an analysis system of learning logs kept in BookLooper system operated in Kyushu University. The present paper overviews the system and shows some "learning log graphs" which represent the learning process of students. The graphs tell which pages a student had difficulties and if the student grasps the thread of course as the teacher expected..
115. Takanori Yamashita, Brendan Flanagan, Yoshifumi Wakata, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Naoki Nakashima, Sachio Hirokawa, Visualization of key factor relation in clinical pathway, 19th International Conference on Knowledge Based and Intelligent Information and Engineering Systems, KES 2015
Procedia Computer Science
, 10.1016/j.procs.2015.08.139, 60, 1, 342-351, 2015.01, [URL], The secondary use of medical data to improve medical care is gaining much attention. We have analyzed electronic clinical pathways for improving the medical process. The analysis of clinical pathways so far has used statistics analysis models, however as issue remains that the order, and multistory spatial and time relations of the each factor could not be analyzed. We constructed an Outcome tree system that shows the greatest significant relation for each factor. The Hip replacement arthroplasty clinical pathway was analyzed by the system, and the outcome variance of the clinical pathway was visualized. The results indicate the path of patient's who have a long hospitalization stay and extracted four critical indicators..
116. Sachio Hirokawa, Tetsuya Nakatoh, Hiroto Nakae, Takahiko Suzuki, Discovery of implicit featurewords of place name, Intelligent Systems Reference Library, 10.1007/978-3-662-47227-9_3, 90, 31-42, 2015, [URL], Individual opinions and experiences are published inWeb as CGM (consumer generated media). A tourism blog which a tourist wrote his experience and impression in a certain area is very helpful information for other tourists. However, a user cannot obtain such precious information without knowing the relation of blog articles and concrete place-names.We paid our attention to the hierarchical structure of place-names. In this paper, we propose the method of connecting related words to the place-name which does not appear explicitly in a blog article paying attention to the hierarchical structure of place-names. From 45,553 blog articles about the Karatsu area in Saga Prefecture, the potential related words about 78 place-names of Saga Prefecture which have not appeared in the blogs were extracted. 4 subjects evaluated that meaningful related words are obtained in 80% or more of the placenames. However, the direct relationships between the place-name and related words was not able to be guessed easily..
117. Tetsuya Nakatoh, Sachio Hirokawa, Extraction of tourism objects from blogs, Intelligent Systems Reference Library, 10.1007/978-3-662-47227-9_4, 90, 43-58, 2015, [URL], Some tourists will write their activities to blogs. Such individual experiences are interesting compared with the official information by tourist agents. Therefore, extracting tourist’s activities from blogs are meaningful. We have attempted to extract tourism objects from tourists’ behavior statistically. In this paper, the objects of the behavior were extracted using dependency analysis, and tourism objects were specified by evaluation with regionality..
118. Toshiro Minami, Kensuke Baba, Sachio Hirokawa, Eriko Amano, A Trichotomic Approach to Approximate Representation of Concepts,
Applied Computing and Information Technology, Studies in Computational Intelligence, 553, 61-75, 2014.12.
119. Shaymaa Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, Comment Data Mining for Student Grade Prediction Considering Differences in Data for Two Classes, The International Association for Computer and Information Science, 15, 2, 12-25, 2014.12.
120. Toshiro Minami, Kensuke Baba, Sachio Hirokawa, Should University Library Collect New Books or Old Books?, Proc. ISAAC2014 & ICACT2014, AACL03, 34-37, 2014.12.
121. Sachio Hirokawa, Brendan Franagan, CHENGJIU YIN, Hiroto Nakae, Vizualization of Relation and Generality of Words in Search Result, Proceedings of the Third Asian Conference on Information Systems, 90-95, 2014.12.
122. Brendan Franagan, CHENGJIU YIN, Sachio Hirokawa, Learning by Search & Log, Proc. ICCE2014, 391-393, 2014.11.
123. Yuichi Ono, Manabu Ishihara, Sachio Hirokawa, Real-time Feedback Systems in a Foreign Language Teaching:
A Case of Presentation Course, Proc. ICCE2014, 779-784, 2014.11.
124. Takanori Yamashita, Yoshifumi Wakata, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Brendan Flanagan, Naoki Nakashim, Sachio Hirokawa, Extraction of Key Factors from Operation Records by
Support Vector Machine and Feature Selection, Indian Journal of Medical Informatics, 8, 70-71, 2014.10.
125. Shaymaa Sorour, Kazumasa Goda, Sachio Hirokawa, Tsunenori Mine, Predicting Students' grades based on free style Comments Data by Artificial Neural Network, Proc. FIE2014, 2014.10.
126. Yuichi Ono, Manabu Ishihara, Sachio Hirokawa, Mitsuo Mamashiro, Real time text-based feedback systems -- From frequency-based feedback
to mindmap feedback in foreign language teaching, Proceedings of IEEE International Conference on Systems, Man and Cybernetics, 2150-2153, 2014.10.
127. Jun Zeng, Min Gao, Junhao Wen, Sachio Hirokawa, A hybrid trust degree model in social network for recommender system, 3rd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2014
Proceedings - 2014 IIAI 3rd International Conference on Advanced Applied Informatics, IIAI-AAI 2014
, 10.1109/IIAI-AAI.2014.19, 37-41, 2014.09, [URL], Recommender system is an effective way to help users to find the required information. In the social network, the recommendation is often from one user to another user. Therefore, it is necessary to determine how the two users trust each other. However, much work has paid more attention to the one-to-one trust relationship but ignored the many-to-one relationship. In this paper, we proposed a hybrid trust degree model to describe how two users trust each other. This model not only considers the direct trust degree and indirect trust degree between the two users, but also considers the group trust degree. The group trust degree describes how a user are trusted by other users in a group. The experiment result shows that hybrid trust degree can reasonably measure and calculate the credit between two users in a group..
128. Xiao Lin, Eisuke Ito, Sachio Hirokawa, Chinese tag analysis for foreign movie contents, 2014 13th IEEE/ACIS International Conference on Computer and Information Science, ICIS 2014 - Proceedings
2014 IEEE/ACIS 13th International Conference on Computer and Information Science, ICIS 2014 - Proceedings
, 10.1109/ICIS.2014.6912126, 163-166, 2014.09, [URL], Consumer Generated Media (CGM) is gaining huge popularity. The authors are particularly interested in the intercultural comprehension of movie contents made in foreign countries. This paper focuses on the website bilibili.tv as a test case to analyze how Japanese movie contents are watched in China. The authors analyze all tags and how foreign tags are introduced and translated into Chinese. They propose a simple statistical method to identify whether a word is a loanword or not, if the word is represented by Chinese characters. They also analyze the trends of tags in bilibili..
129. Takuya Hirao, Takahiko Suzuki, Koki Miyata, Sachio Hirokawa, Detection of misplacement of synonyms in the Japanese WordNet, 3rd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2014
Proceedings - 2014 IIAI 3rd International Conference on Advanced Applied Informatics, IIAI-AAI 2014
, 10.1109/IIAI-AAI.2014.18, 31-36, 2014.09, [URL], Lexical database the Japanese WordNet is a useful tool in natural language processing. However, it is officially announced that the Japanese WordNet contains 5% errors. In this paper, we classify errors in the Japanese WordNet and discuss error detection methods..
130. Sachio Hirokawa, Tetsuya Nakatoh, Hiroto Nakae, Takahiko Suzuki, Discovery of implicit feature words of place name, 3rd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2014
Proceedings - 2014 IIAI 3rd International Conference on Advanced Applied Informatics, IIAI-AAI 2014
, 10.1109/IIAI-AAI.2014.122, 561-566, 2014.09, [URL], Individual opinions and experiences are published in Web as CGM (consumer generated media). A tourism blog which a tourist wrote his experience and impression in a certain area is very helpful information for other tourists. However, a user cannot obtain such precious information without knowing the relation of blog articles and concrete place-names. We paid our attention to the hierarchical structure of place-names. In this paper, we propose the method of connecting related words to the place-name which does not appear explicitly in a blog article paying attention to the hierarchical structure of place-names. From from 45,553 blog articles about the Karatsu area in Saga Prefecture, the potential related words about 78 place-names of Saga Prefecture which have not appeared in the blogs were extracted. 4 subjects evaluated that meaningful related words are obtained in 80% or more of the place-names. However, the direct relationships between the place-name and related words was not able to be guessed easily..
131. Sachio Hirokawa, Message from conference general chair, Quaternary International, 10.1109/IIAI-AAI.2014.5, xx, 2014.09, [URL].
132. Jun Zeng, Min Gao, Junhao Wen, Sachio Hirokawa, A Hybrid Trust Degree Model in Social Network, Proc. AAI2014, 37-41, 2014.08.
133. Brendan Franagan, CHENGJIU YIN, Takahiko Suzuki, Sachio Hirokawa, Classification and clustering English Writing Errors based on Native language, Proc. AAI2014, 318-324, 2014.08.
134. Shaymaa Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, Comments data mining for evaluating student's performance, Proc. AAI2014, 25-30, 2014.08.
135. Takanori Yamashita, Yoshifumi Wakata, Naoki Nakashim, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Brendan Flanagan, Sachio Hirokawa, Construction of Dominant Factor Presumption Model for
Postoperative Hospital Days from Operation, Proc. AAI2014, 19-24, 2014.08.
136. Takuya Hirao, Koki Miyata, Takahiko Suzuki, Sachio Hirokawa, Detection of Misplacement of Synonyms in the Japanese WordNet, Proc. AAI2014, 31-36, 2014.08.
137. Sachio Hirokawa, Tetsuya Nakatoh, Hiroto Nakae, Takahiko Suzuki, Discovery of Implicit Feature Words of Place Name, Proc. AAI2014, 561-566, 2014.08.
138. Yuichi Ono, Sachio Hirokawa, Manabu Ishihara, Mitsuo Yamashiro, Implementation and evaluation of real time qualitative feedback
systems in a foreign language presentation course, Proc. AAI2014, 372-376, 2014.08.
139. Sachio Hirokawa, Brendan Flanagan, Takahiko Suzuki, CHENGJIU YIN, Learning Winespeak from Mind Map of Wine Blogs, Proc. HIMI 2014, Part II, LNCS 8522, 383-393, 2014.07.
140. Sachio Hirokawa, Emi Ishita, Non-Topical Classification of Healthcare Information on the Web, Frontiers in Artificial Intelligence and Applications, 383-393, 2014.07, The present paper collected the asthma related 4,762 Web pages from 1,759 sites using 6 queries. Each site is manually categorized by the standard topics of description and information dissemination, diary and idle talk and Q&A. By careful analysis, it turned out that the pages can be classified in non-topical categories such as “reading level”, “objectivity/subjectivity” and “reliability”. The manually assigned labels of non-topical categories are then used as learning data to apply SVM (support machine vector). The prediction performance (F-measure) were below 50% with the naive application of SVM. However, the prediction performance was improved over 50% by feature selection except for reading level..
141. Takanori Yamashita, Yoshifumi Wakata, Naoki Nakashim, Sachio Hirokawa, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Extraction of Determinants of Postoperative Length of Stay from Operation Records, Proceedings of 2014 IEEE Workshop on Electronics, Computer and Applications, 8, 822-827, 2014.05.
142. Shaymaa Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, Efficiency of LSA and K-means in perdicting studnt's academic performance based on comments data, Proc. CSEDU2014, 43-52, 2014.04.
143. Jun Zeng, Brendan Flanagan, Sachio Hirokawa, Eisuke Ito, A Web Page Segmentation Approach Using Visual Semantics, IEICE Trans. Inf. & Sys., E97-D, 2, 223-230, 2014.02, Web page segmentation has a variety of benefits and potential web
applications. The early techniques of web page segmentation are mainly
based on machine learning algorithms and rule-based heuristics,
which cannot be used for large - scale page segmentation. In this
paper, we propose a formulated page segmentation method using visual
semantics. Instead of analyzing the visual cues of web pages, this
method utilizes three measures to formalize the visual semantics:
layout tree is used to recognize the visual similar blocks; seam
degree is used to describe how neatly the blocks are arranged; content
similarity is used to describe the distance between the blocks. A
comparison experiment w as done u sing the VIPS algorithm as the
baseline. The experimental results show that our method can divide a
Web page into appropriate semantic segments..
144. Chengjiu Yin, Yoshiyuki Tabata, Sachio Hirokawa, A "milky way research trend" system for survey of scientific literature, 2012 International Workshops on Web-Based Learning, ICWL 2012 - KMEL, SciLearn, and CCSTED
New Horizons in Web Based Learning - ICWL 2011 International Workshops KMEL, ELSM, and SPeL and ICWL 2012 International Workshops KMEL, SciLearn, and CCSTED, Revised Selected Papers
, 10.1007/978-3-662-43454-3_10, 7697 LNCS, 90-99, 2014.01, [URL], Research trend survey is an essential preliminary step for any academic researches, but many beginning researchers have difficulty because they are still foreign to appropriate keywords in his/her research field. We constructed a support system for research trend surveys not only to accelerate the preliminary step but also to let students have a better grips of trend progresses and keyword transitions. Our system assumes a fair amount of data accumulation, for which we employed KAKEN database excerpts, but does not assume manual keyword registration or any other heuristic preprocesses: with an associative search module, it dynamically searches relevant words that are frequently used in the targeted academic field and gives users effective visualizations to understand trend transitions. Preliminary evaluations suggest that the trend transitions that our system presents are effective for trend surveys..
145. Jun Zeng, Brendan Flanagan, Sachio Hirokawa, Eisuke Ito, A web page segmentation approach using visual semantics, IEICE Transactions on Information and Systems, 10.1587/transinf.E97.D.223, E97-D, 2, 223-230, 2014.01, [URL], Web page segmentation has a variety of benefits and potential web applications. Early techniques of web page segmentation are mainly based on machine learning algorithms and rule-based heuristics, which cannot be used for large-scale page segmentation. In this paper, we propose a formulated page segmentation method using visual semantics. Instead of analyzing the visual cues of web pages, this method utilizes three measures to formulate the visual semantics: layout tree is used to recognize the visual similar blocks; seam degree is used to describe how neatly the blocks are arranged; content similarity is used to describe the content coherent degree between blocks. A comparison experiment was done using the VIPS algorithm as a baseline. Experiment results show that the proposed method can divide a Web page into appropriate semantic segments..
146. Brendan Flanagan, Chengjiu Yin, Takahiko Suzuki, Sachio Hirokawa, Classification and clustering English writing errors based on native language, 3rd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2014
Proceedings - 2014 IIAI 3rd International Conference on Advanced Applied Informatics, IIAI-AAI 2014
, 10.1109/IIAI-AAI.2014.72, 318-323, 2014.01, [URL], It is important for language learners to determine and reflect on their writing errors in order to overcome weaknesses. Each language learner has their own unique writing error characteristics and therefore has different learning needs. In this paper, we analyze the writing errors of foreign language learners on the language learning SNS website Lang-8 to investigate the characteristics of errors by native language. 142,465 sentences were collected from Lang-8 for analysis. For each native language, the predicted scores of 15 error categories from SVM machine learning models are used as a vector representation of each sentence. These score vectors are then clustered to determine error co-occurrence within the same sentence. The results were then analyzed to determine the error characteristics of different native languages..
147. Shaymaa E. Sorour, Tsunenori Mine, Kazumasa Godaz, Sachio Hirokawa, Comments data mining for evaluating student's performance, 3rd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2014
Proceedings - 2014 IIAI 3rd International Conference on Advanced Applied Informatics, IIAI-AAI 2014
, 10.1109/IIAI-AAI.2014.17, 25-30, 2014.01, [URL], The present study proposes prediction approaches of student's grade based on their comments data. Students describe their learning attitudes, tendencies and behaviors by writing their comments freely after each lesson. The main difficulty of this research is to predict students' performance by separately using two class data in each lesson. Although students learn the same subject, there exist differences between the comments in the two classes. The proposed methods basically employ latent semantic analysis (LSA) and two types of machine learning technique: SVM (support vector machine) and ANN (artificial neural network) for predicting students' final results in four grades of S, A, B and C. Moreover, an overlap method was proposed to improve the accuracy prediction results, the method allows to accept two grades for one mark to get the correct relation between LSA results and students' grades. The proposed methods achieve 50.7% and 48.7% prediction accuracy of students' grades by SVM and ANN, respectively. To this end, the results of this study reported models of students' academic performance predictors that are valuable sources of understanding students' behavior and giving feedback to them so that we can improve their learning activities..
148. Takanori Yamashita, Yoshifumi Wakata, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Brendan Flanagan, Naoki Nakashima, Sachio Hirokawa, Construction of dominant factor presumption model for postoperative hospital days from operation records, 3rd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2014
Proceedings - 2014 IIAI 3rd International Conference on Advanced Applied Informatics, IIAI-AAI 2014
, 10.1109/IIAI-AAI.2014.16, 19-24, 2014.01, [URL], The secondary use of clinical text data to improve the quality and the efficiency of medical care is gaining much attention. However, there are few previous researches that have given feedback to clinical situations. The present paper analyzes the words that appear in operation records to predict the postoperative length of stay. SVM (support vector machine) and feature selection are applied to predict if a stay is longer than the standard length of 25 days. It was confirmed that with less than 20 feature words we can predict if a stay is longer or not with almost the optimal prediction performance..
149. Shaymaa E. Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, Efficiency of LSA and K-means in predicting students' academic performance based on their comments data, 6th International Conference on Computer Supported Education, CSEDU 2014
CSEDU 2014 - Proceedings of the 6th International Conference on Computer Supported Education
, 1, 63-74, 2014.01, Predicting students' academic performance has long been an important research topic in many academic disciplines. The prediction will help the tutors identify the weak students and help them score better marks; these steps were taken to improve the performance of the students. The present study uses free style comments written by students after each lesson. These comments reflect their learning attitudes to the lesson, understanding of subjects, difficulties to learn, and learning activities in the classroom. (Goda and Mine, 2011) proposed PCN method to estimate students' learning situations from their comments freely written by themselves. This paper uses C (Current) method from the PCN method. The C method only uses comments with C item that focuses on students' understanding and achievements during the class period. The aims of this study are, by applying the method to the students' comments, to clarify relationships between student's behaviour and their success, and to develop a model of students' performance predictors. To this end, we use Latent Semantic Analyses (LSA) and K-means clustering techniques. The results of this study reported a model of students' academic performance predictors by analysing their comment data as variables of predictors..
150. Takanori Yamashita, Yoshifumi Wakata, Naoki Nakashima, Sachio Hirokawa, Satoshi Hamai, Yasuharu Nakashima, Yukihide Iwamoto, Extraction of determinants of postoperative length of stay from operation records, 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014
Proceedings - 2014 IEEE Workshop on Electronics, Computer and Applications, IWECA 2014
, 10.1109/IWECA.2014.6845748, 822-827, 2014.01, [URL], Secondary use of clinical text data are gaining much attention in improving the quality and the efficiency of medical treatment. Although there is some case studies of medical-examination text data, there are not many examples fed back to the medical-examination spot. The present paper analyses the operation records of total hip arthroplasty. We extracted feature words that characterize the two peaks which appeared in distribution of postoperative hospital days using SVM (support vector machine) and FS (feature selection). The models gained by optimal FS attained 60% accuracy as prediction performance. We applied logistic regression analysis to estimate postoperative length of stay from the extracted feature words. Most words were not statistically significant except two words..
151. Yuichi Ono, Manabu Ishihara, Sachio Hirokawa, Mitsuo Yamashiro, Implementation and evaluation of real time qualitative feedback systems in a foreign language presentation course, 3rd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2014
Proceedings - 2014 IIAI 3rd International Conference on Advanced Applied Informatics, IIAI-AAI 2014
, 10.1109/IIAI-AAI.2014.83, 372-376, 2014.01, [URL], This paper attempts to evaluate our new real time qualitative feedback system and its implementation. It was suggested in our earlier research that text-based instant feedback has an advantage of providing learners with opportunities to become aware of new discoveries from the feedback from the audience. We start the discussion with the consideration of some issues around the new system, comparison with two traditional approaches, Clicker approach and Forum approach, cost on learners, and their motivation for the next presentation. We carried out experiment studies and prove that our approach is superior to the traditional Forum and quantitative approach, showing the results of our research..
152. Chengjiu Yin, Brendan Flanagan, Sachio Hirokawa, Learning by "search & log", 22nd International Conference on Computers in Education, ICCE 2014
Proceedings of the 22nd International Conference on Computers in Education, ICCE 2014
, 391-393, 2014.01, Although previous research has demonstrated the benefits of the "learning by searching" strategy, there is a new problem which is how to measure and analyze the effectiveness of "Learning by Searching" behaviors. In this paper, by using the record of the students' learning history, we have proposed a SNSearch system to analyze student web-searching behaviors of "Learning by Searching"..
153. Sachio Hirokawa, Brendan Flanagan, Takahiko Suzuki, Chengjiu Yin, Learning winespeak from mind map of wine blogs, 16th International Conference on Human Interface and the Management of Information: Information and Knowledge Design and Evaluation, HCI International 2014
Human Interface and the Management of Information
Information and Knowledge in Applications and Services - 16th International Conference, HCI International 2014, Proceedings
, 10.1007/978-3-319-07863-2_37, 383-393, 2014.01, [URL], When faced with complex situations, it can often be hard to put into words and accurately express it appropriately. This becomes increasingly difficult when specialist expressions are required that are not used in everyday language. The problem is faced when trying to express in words to another person the wine that you just drank, or a wine that you want to drink to a waiter at a restaurant or shop assistant. It requires the expression in words of numerous senses including complex flavors, smells, colors, and personal emotion that is felt. These expressions are often subjective, with different people having using different expressions for the same wine. In this paper, we propose the use of wine related expressions collected from the internet and clustered to generate mind maps..
154. Toshiro Minami, Sachio Hirokawa, Kensuke Baba, Eriko Amano, A trichotomic approach to concept capture and representation
With its application to library data mining, Studies in Computational Intelligence, 10.1007/978-3-319-05717-0_5, 553, 61-75, 2014, [URL], The aim of this chapter is twofold. Firstly, we propose a method of specifying the concept that is too hard to describe in an exact way by a word or a phrase, by setting up the “relative distances" from three key concepts; which we call a trichotomic approach to concept capture and representation, or description, in an approximate means. It is important and interesting that we can choose not only the key words but also other three “keys" such as patrons, books, concepts, objects or others. Then we arrange the objects of study according to the relative distances from these three keys, and investigate how these objects are distributed. Secondly, we demonstrate the usefulness of trichotomic approach through a couple of case studies applied to library’s loan record analysis. In these case studies, we discuss and compare the methods of choosing three keys, then we show how the trichotomic representation method is applied to the real data analysis. From these case studies, we are convinced of its high potential and importance as a visualization tool of the results of data analysis in general..
155. Jun Zeng, Brendan Flanagan, Qingyu Xiong, Junhao Wen, Sachio Hirokawa, A web page segmentation approach using seam degree and content similarity, Studies in Computational Intelligence, 10.1007/978-3-319-05717-0_7, 553, 91-103, 2014, [URL], Page segmentation has received great attention in recent years. However, most research has been based on some pre-defined heuristics or visual cues which may be not suitable for large-scale page segmentation. In this chapter, we proposed two parameters: seam degree and content similarity, to indicate the coherent degree of a page block. Instead of analyzing pre-defined heuristics or visual cues, our method utilizes the visual and content features to determine whether a page block should be divided into smaller blocks. We also proposed a principled page segmentation method using these two parameters. An experiment was conducted to determine the relationship between the two parameters and the number of segment results. The empirical results also show that our segmentation method can effectively segment a page into different semantic parts..
156. Sachio Hirokawa, Emi Ishita, Non-topical classification of healthcare information on the web, Smart Digital Futures 2014, 10.3233/978-1-61499-405-3-237, 262, 237-247, 2014, [URL], The present paper collected the asthma related 4,762 Web pages from 1,759 sites using 6 queries. Each site is manually categorized by the standard topics of description and information dissemination, diary and idle talk and Q&A. By careful analysis, it turned out that the pages can be classified in non-topical categories such as 'reading level', 'objectivity/subjectivity' and 'reliability'. The manually assigned labels of non-topical categories are then used as learning data to apply SVM (support machine vector). The prediction performance (F-measure) were below 50% with the naive application of SVM. However, the prediction performance was improved over 50% by feature selection except for reading level..
157. Shaymaa E. Sorour, Tsunenori Mine, Kazumasa Goda, Sachio Hirokawa, Prediction of students' grades based on free-style comments data, 13th International Conference on Advances in Web-Based Learning, ICWL 2014
Advances in Web-Based Learning, ICWL 2014 - 13th International Conference, Proceedings
, 10.1007/978-3-319-09635-3_15, 8613 LNCS, 142-151, 2014, [URL], In this paper we propose a new approach based on text mining technique to predict student's performance using LSA (latent semantic analysis) and K-means clustering method. The present study uses free style comments written by students after each lesson. Since the potentials of these comments can reflect students' learning attitudes, understanding and difficulties to the lessons, they enable teachers to grasp the tendencies of students' learning activities.To improve this basic approach, overlap method and similarity measuring technique are proposed. We conducted experiments to validate our proposed methods. The experimental results illustrated that prediction accuracy was 73.6% after applying the overlap method and that was 78.5% by adding the similarity measuring..
158. Yuichi Ono, Sachio Hirokawa, Manabu Ishihara, Mitsuo Yamashiro, Real time text-based feedback systems
From frequency-based feedback to mindmap feedback in foreign language teaching, Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, 10.1109/smc.2014.6974240, 2014-January, January, 2150-2153, 2014, [URL], This paper is concerned with the implementation of text-based feedback system. Quantitative feedback approach like "Clicker" has been common to enhance participants' involvement in the classroom and improve the quality of interactivity. On the other hand, there is a common approach to text-based feedback like the use of social networks and bulletin-boards installed in Learning Management Systems. After reviewing limitations of these two approaches, we would like to propose a real time text-based feedback system with focus on the frequency of keywords used by the audience. This paper further proposes "Mind Mapping" feedback approach on the basis of the dictionary used in the system. It is suggested that the implementation of the system in foreign language presentation course in Japan had an effect on awareness and motivation of Japanese learners of English..
159. Yuichi Ono, Manabu Ishihara, Sachio Hirokawa, Mitsuo Yamashiro, Real-time feedback systems in a foreign language teaching
A case of presentation course, 22nd International Conference on Computers in Education, ICCE 2014
Proceedings of the 22nd International Conference on Computers in Education, ICCE 2014
, 779-784, 2014, This paper is concerned with a new type of real-time feedback system in a classroom based on the text data collected from the audience. After reviewing two traditional approaches to real-time feedback; the Clicker Approach and the Forum Approach, it will be suggested that either of them is insufficient as a tool to motivate the learners in a case of presentation course. Instead, we propose two systems on the basis of text-mining technique to compensate for these insuffiencies. The first is a "Keyword and Frequency" system and the other is "Mind-mapping" system. In this paper, we describe the details of the systems. By being presented the keywords and data on frequency, the presenters can easily understand about the general feedback tendencies. In addition, the mind-map picture gives the presenters the opportunities of promoting a new awareness, various kinds of discoveries, and a deeper reflection about their works. Totally, our system can be incorporated into Learning Management System (LMS), and it has a large potential for further use in a distant learning environment to capture an overall reaction from the audience all over the world..
160. Brendan Flanagan, CHENGJIU YIN, Yohei Inokuchi, Sachio Hirokawa, Supporting Interpersonal Communication Using Mind-Map, Journal of Information and Systems in Education, 2013.12, Engaging in initial communication interactions with an unknown partner
for the first time can be a daunting task. Participants often use
several different strategies, of which an active strategy involves
getting to know their communication partners through asking other
people. However this situation might not always be possible. To
support situation when this strategies isn’t available, we propose a
method to extract keywords from a user’s comments on SNS sites and
use these to represent their interests and activities. Keywords are
visualized as a mind map for use as a communication tool when engaging
in interpersonal communications. The use of this method is
de monstrated and evaluated in three examples that were created from
real world data collected from Twitter.
.
161. Makoto Okada, Sachio Hirokawa, Kiyota Hashimoto, Analysis of opinions from questionnaire surveys of farming candidates using cross tabulation system, Intelligent Interactive Multimedia Systems and Services. Proceedings of the 6th International Conference on Intelligent Interactive Multimedia Systems and Services (IIMSS2013), 10.3233/978-1-61499-262-2-184, 184-194, 2013.12, [URL], In Japan, agricultural activation poses a big problem and the increase in new-entrants-to-agriculture persons is expected. This paper analyzes questionnaire surveys of those who want to start working in agriculture to figure out the difficulties and anxiety they have. The questionnaires consist of the categorical data, such as status of their willingness to be a farmer, their age and sex and the free texts that are written by person who answered several sets of questions. Categories and free text question items are considered as viewpoints in this paper. Two viewpoints are selected to construct a cross tabulation of a search result. The number of the questionnaires that matches the two conditions is displayed in the cell of the cross table. All possible cross tables can be shown as a result. A specified cross table is shown if the user determines a pair of viewpoints. This paper reports case studies which are hard to achieve by simple keyword search..
162. Eisuke Ito, Takahiro Urakawa, Brendan Flanagan, Sachio Hirokawa, Keywords frequency trend analysis of online novels, 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
Proceedings - 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
, 10.1109/IIAI-AAI.2013.92, 68-73, 2013.12, [URL], The authors are interested in online novel services as a user-generated media on the Web. A large number of novels are being uploaded, and a few novels become major. Novel writers like to create novel of current popular genre, then current popular genre words may frequently appear. In this paper, the authors apply the time series analysis to the keywords words given to an online novel by the creator. The authors construct a trend analysis tool. The tool not only shows the trend of posted query word(s), but also shows the trends of similar terms. This paper describes the trend analysis system, the used data, and some interesting analyses..
163. Yasuhiro Yamada, Terutaka Tansho, Sachio Hirokawa, Proposal of a matching system for companies and researchers using patents and scientific papers, 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
Proceedings - 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
, 10.1109/IIAI-AAI.2013.55, 397-398, 2013.12, [URL], For open innovation and industry-university cooperation, it is important for companies and university researchers to find suitable partners to ensure collaborative development. However, it is difficult to find partners via traditional information retrieval systems since they only output a list of patent and scientific paper related to an input query. This paper proposes a system that matches companies with university researchers, and vice versa, by using patents and scientific papers..
164. Eisuke Ito, Brendan Flanagan, CHENGJIU YIN, Tetsuya Nakatoh, Sachio Hirokawa, A study of search engine exercise using VM on a private cloud, Proc. ICCE2013, 2013.11.
165. Brendan Flanagan, CHENGJIU YIN, Kiyota Hashimoto, Sachio Hirokawa, Yoshiyuki Tabata, An Automated Method to Generate e-Learning Quizzes from Online Language Learner Writing, International Journal of Distance Education Technologies, 2013.11, In this paper, the entries of Lang-8, which is a SNS site for learning
and practicing foreign languages, were analyzed and found to contain
similar rates of errors for most error categories reported in previous
research. These similarly rated errors were then processed using an
algorithm to determine corrections suggested by a native
speaker. Subject matter experts then evaluated the processed sentences
to determine the quality in relation to use in tests or exams for
language learners. The method describes the automatic generation of
multiple choice and fill-in-the-blanks quizzes using the writings of
language learners on public web based learning sites, in order to
support learner reflection on corrections and practicing past errors
to overcome problems.
.
166. Makoto Okada, Sachio Hirokawa, Kiyota Hashimoto, An Investigation of a Method to Extract Japanese Start-Farming
Problems using Comparison News Articles and Questionnaire Surveys, Proc. AIT2013, 2013.11.
167. Brendan Flanagan, Jane Yin-Kim YAU, Sachio Hirokawa, Yoshiyuki Tabata, An SNS based Literature Review System for conducting a Research Survey, Proc. ICCE2013, 2013.11, It is necessary to perform a literature review before starting a new
research project . However, many students do not know the procedur es
of performing a literature review . In this paper, based on the p
rofessional experiences and opinions of expert researchers , we
describe a n SNS - based literature review system to help students
conduct research survey s . This system includes two search engi nes,
one is a n article search engine , which can help students conduct
research survey s , and the other is a logging search engine , which
allows students to learn from each other via their logs and share
experience with other students . User models of the sys tem as well
as its functions are presented..
168. Brendan Flanagan, CHENGJIU YIN, Kiyota Hashimoto, Sachio Hirokawa, Clustering English Writing Errors based on Error Category Prediction, Proc. ISEEE2013, 65-70, 2013.11, It is important for language learners to determine and reflect on
their writing errors in order to overcome weaknesses. Each language
learner has their own unique writing error characteristics and
therefore has different learning needs. The present paper applies SVM
machine learning to the writings of English language learners on the
language learning SNS website Lang-8 that have been manually
classified into 15 error categories to determine the errors in a
sentence. Feature selection was used to improve the performance of
the resulting classifier model.
.
169. Koki Miyata, Takahiko Suzuki, Sachio Hirokawa, Difficulty and Ambiguity of Verbs -Analysis based on Synsets in Japanese WordNet-, Proc. AIT2013, 2013.11.
170. Brendan Flanagan, Jun Zeng, Brendan Flanagan, Sachio Hirokawa, Extraction of Informative Blocks from Deep Web Page Using Similar Layout Feature, International Journal of Advancements in Computing Technology, 5, 9, 316-324, 2013.11, Due to the explosive growth and popularity of the deep web,
information extraction from deep web page has gained more and more
attention. However, the HTML structure of web page has become more
complicated, making it difficult to recognize target content by only
analyzing the HTML source code. In this paper, we propose a method to
extract the informative blocks from a deep web using the layout
feature. We consider the visual rectangular region of an HTML element
as a visual block in web page. We transform the elements' layout of a
visual block into a layout tree. By calculating the similarity of
layout trees, we cluster the visual blocks that have similar layout
feature. Finally, the cluster which has the largest area is extracted
as the informative block cluster. The experiment results show that
this method is optimal when the threshold of layout tree similarity is
0.4..
171. Sachio Hirokawa, Integration of Sentence Based and Document Based Text Mining of Securities Reports for Prediction of Growth Rate of Operating Income, Proc. IRCITCS 2013, 132-144, Integration of Sentence Based and Document Based Text Mining of Securities Reports for Prediction of Growth Rate of Operating Income, 2013.11.
172. Eisuke Ito, Sachio Hirokawa, Keyword Relation Analysis Using Concept Graph Toward Automatic Categorization of Online Novels, Proc. AIT2013, 2013.11.
173. CHENGJIU YIN, Han-Yu Sung, Gwo-Jen Hwang, Sachio Hirokawa, Hui-Chun Chu, Brendan Flanagan, Yoshiyuki Tabata, Learning by Searching: A Learning Environment that Provides Searching and Analysis Facilities for Supporting Trend Analysis Activities, Journal of Educational Technology & Society, 283-300, 2013.11, With the popularity of the Internet, online searching is
becoming an important part of learning. In this paper, based on the
"Learning by Searching theory, a learning environment is developed,
which includes a search engine to assist students in recognizing the
progression of trends and keyword transitions for specific domains. To
efficiently support research trend surveys, an automatic data
accumulation and classification approach is proposed to construct the
database excerpts instead of manual keyword registration or any other
heuristic preprocesses. With an associative search module, the search
engine dynamically searches for relevant words that are frequently
used in the targeted academic field, and provides learners with
effective visualizations to understand the trend transitions. An
experiment has been conducted on a college information management
course to show the effectiveness of the proposed approach. The
experiment results show that the students who learned with the new
approach had significantly better learning performance in terms of
recognizing the trend transitions of the targeted issues than those
who learned with conventional search engines..
174. Brendan Flanagan, Takahiko Suzuki, Jun Zeng, CHENGJIU YIN, Toshihiko Sakai, Kiyota Hashimoto, Sachio Hirokawa, Peer Knowledge Assisted Search Using Community Search Logs, International Journal of Ditigal Information and Wireless Communications(IJDIWC) , 3, 2, 1-8, 2013.11.
175. Kazumasa Goda, Sachio Hirokawa, Tsunenori Mine, Correlation of grade prediction performance and validity of self-evaluation comments, 2013 13th ACM SIGITE Annual Conference on Information Technology Education, SIGITE 2013
SIGITE 2013 - Proceedings of the 2013 ACM SIGITE Annual Conference on Information Technology Education
, 10.1145/2512276.2512294, 35-42, 2013.11, [URL], To grasp a student's lesson attitude and learning situation and to give a feed back to each student are educational foundations. Goda et al. proposed the PCN method to estimate a learning situation from a comment freely written by students[6, 7]. The PCN method categorizes comments into three items of P (previous), C(current) and N(next). They pointed out a correlation between the student's final results and the validity of a descriptive content of item C, that is something related to understanding of the lesson and learning attitudes to the lesson. However, a problem left in their work is the badness of performance in prediction for upper grade students. This paper proposes two manners of utilization of PCN scores: the validity level determination for assessment, and for prediction performance of students' final grades. In order to validate the proposed manners of utilization, we conducted two experiments. First, we employed multiple regression analysis to calculate PCN scores that determine the validity level with respect to each viewpoint. Students who wrote comments with a high PCN score are considered as those who describe their learning attitude appropriately. We also applied a machine learning method SVM (support vector machine) to students' comments for predicting their final results in five grades of S, A, B, C and D. Experimental results illustrated that as comments of students get higher PCN scores, the prediction performance of the students' grades becomes higher..
176. Kazumasa Goda, Sachio Hirokawa, Tsunenori Mine, Automated Evaluation of Student Comments on Their Learning Behavior, Proc. ICWL2013, 131-140, 2013.10, Learning comments are valuable source of interpreting student status
of understanding. The PCN method introduced in [Gouda2011] analyzes
the attitude of a student from a view point of time series. Each
sentence of a comment are manually classified as one of P,C,N or O
sentence. P(previous) indicates the learning activity before the
classtime, C(current) represents the understanding and achievement
during the classtime, and N(next) means the learning activity plan
until next class. The present paper applies SVM(Support Vecotor
Machine) to predict the category to which given sentence
belongs. Empirical evaluation using 4,086 sentences was conducted. By
selecting feature words of each category, the prediction performance
was satisfactory with F-measures 0.8203, 0.7352, 0.8416 and 0.8612 for
P,C,N and O respectively.
.
177. Kazumasa Goda, Sachio Hirokawa, Tsunenori Mine, Correlation of Grade Prediction Performance and Validity of Self-Evaluation Comments, Prof. ACM SIGITE2013, 35-42, 2013.10, To grasp a student's lesson attitude and learning situation and to
give a feed back for each student are educational foundations. Goda
et.al. 2011 proposed the PCN method to presume a learning situation
from a comment freely written by students. The PCN method categorizes
comments into three items of P(previous), C(current) and N(next).
Item P is learning activities for preparation of a lesson. Item C is
understanding of the lesson and learning attitudes to the lesson. Item
N is the learning plan and goal by the next lesson. They pointed out a
correlation between the student's final results and the validity of a
descriptive content of item C.
A problem left in Goda et. al.2011 is the badness of performance in prediction
for upper grade students. One of the reason is the
difficulty in assessmenting the validity by human. Another reason is
the diversity of quality of freely written comments. Some students
wrote their comments that have nothing to do with their understanding
or learning attitudes in item C.
The present paper applied multiple regression analysis to calculate
the PCN scores that determine the validity level with respect to each
viewpoint. The students who wrote comments with high PCN score are
considered as those who describe their learning attitude appropriately.
The present paper applied a machine learning method SVM (support
vector machine) to student comments for predicting the students' final
result in five grades of S,A,B,C and D. It is confirmed that the
prediction performance of student grade is high for the students with
high PCN scores.
.
178. Kazumasa Goda, Sachio Hirokawa, Tsunenori Mine, Automated evaluation of student comments on their learning behavior, 12th International Conference on Web-based Learning, ICWL 2013
Lecture Notes in Computer Science
, 10.1007/978-3-642-41175-5_14, 8167 LNCS, 131-140, 2013.10, [URL], Learning comments are valuable sources of interpreting student status of understanding. The PCN method introduced in [Gouda2011] analyzes the attitudes of a student from a view point of time series. Each sentence of a comment is manually classified as one of P,C,N or O sentence. P(previous) indicates learning activities before the classtime, C(current) represents understanding or achievements during the classtime, and N(next) means a learning activity plan or goal until next class. The present paper applies SVM(Support Vecotor Machine) to predict the category to which a given sentence belongs. Empirical evaluation using 4,086 sentences was conducted. By selecting feature words of each category, the prediction performance was satisfactory with F-measures 0.8203, 0.7352, 0.8416 and 0.8612 for P,C,N and O respectively..
179. CHENGJIU YIN, Brendan Flanagan, Sachio Hirokawa, Yoshiyuki Tabata, Build A Search Engine To Support Doing Research Surveys On SNS, Proc. LTLE2013, 2013.09, It is very important for any academic researches to do research
surveys. In this paper, we proposed an SNS based search engine to
support doing research surveys, called SNSearch. Our SNSearch system
not only supports analysis of research trends, but also offers
learners opportunities to reflect on their search behaviors. By
browsing the experts’ survey history, students can learn search
skills..
180. Tetsuya Nakatoh, Sachio Hirokawa, Evaluation of Tourism Resources Extraction based on Japanese Dependency Analysis, Proc. ESKM 2013, 100-1003, 2013.09.
181. Eisuke Ito, T. Urakawa, Brendan Flanagan, Sachio Hirokawa, Keywords Frequency Trend Analysis of Online Novels, Proc. ESKM 2013, 68-73, 2013.09.
182. Tetsuya Nakatoh, Hirofumi Amano, Sachio Hirokawa, Prediction of Growth Rate of Operating Income using Securities Reports, Proc. ESKM 2013, 2013.09, Blog articles by tourists contain interesting and
personal experiences of where and how they have gone, what they
have done and what they thought. Such individual experiences
are helpful in many cases compared to the general and official
information about the tourist resort by tourist agents. However,
it is not easy to choose related articles and to extract still
more nearly required information from these unsorted blog
articles. This paper proposes a technique of feature extraction
by dependency analysis of verbs and objects in those sentences
that describe tourist’s behavior. This paper applied the method
to 70,352 blog articles on Kyushu area and reports some analysis
on "where and what did they eat" as case studies.
.
183. Jun Zeng, Brendan Flanagan, Qingyu Xiong, Junhao Wen, Sachio Hirokawa, Proposal of Seam Degree and Content Similarity for Web Page Segmentation, Proc. ESKM 2013, 2013.09, Page segmentation has received great attention in recent
years. However, most research has been based on some pre-defined
heuristics or visual cues which may be not suitable for large-scale
page segmentation. In this paper, we proposed two parameters: seam
degree and content similarity, to indicate the coherent degree of a
page block. Instead of analyzing pre-defined heuristics or visual
cues, our method utilizes the visual and content features to determine
whether a page block should be divided into smaller blocks. We also
proposed a principled page segmentation method using these two
parameters. An experiment was conducted to determine the relationship
between the two parameters and the number of segment results. The
empirical results also show that our segmentation method can
effectively segment a page into different semantic parts.
.
184. Yasuhiro Yamada, Terutaka Tansho, Sachio Hirokawa, Proposal of a Matching System for Companies and Researchers Using Patents and Scientific Papers, Proc. ESKM 2013, 397-398, 2013.09.
185. Brendan Flanagan, Yohei Inokuchi, CHENGJIU YIN, Sachio Hirokawa, Using Automatically Generated Mind Maps to Promote Initial Communication, Proc. AECT International Conference on the Frontier in e-Learning Research 201, 332-333, 2013.08, In communication, it is important to get to know the common
interests of each other. This paper proposes a method to extract
keywords that represent the interests of individuals from their
comments on SNS sites. Those keywords are visualized as a mind map,
which can be used as a communication tool. The effectiveness of the
maps is evaluated in three examples.
.
186. Hongjie ZHAI, Makoto HARAGUCHI, Yoshiaki OKUBO, Kiyota HASHIMOTO, Sachio Hirokawa, Shifting Concepts to Their Associative Concepts via Bridge, Proc. MLDM2013, Springer LNCS 7988, 586-600, 2013.07, This paper presents a pair of formal concept search procedures to find
associative connection of concepts via bridge concepts. A bridge is a
generalization of a sub-concept of an initial concept. The initial
concept is then shifted to other target concepts which are
conditionally similar to the initial one within the extent of
bridge. A procedure for mining target concepts under the conditional
similarity with respect to the bridge is presented based on an
object-feature incident relation. Such a bridge concept is constructed
in the concept lattice of person-feature incident relation. The latter
incident relation is defined by aggregating the former
document-feature relation to have more condensed relation, while
keeping the variation of possible candidate bridges. Some heuristic
rule, named Mediator Heuristics, is furthermore introduced to reflect
user’s interests and intention. The pair of these two procedures
provides an efficient method for shifting initial concepts to target
ones via some bridges. We show their usefulness by applying them to
Twitter data .
.
187. Li Jian, Cheng Jiu Yin, Sachio Hirokawa, Yoshiyuki Tabata, Modified differential evolution for tension/Compression string design problem, 2013 International Conference on MEMS and Mechanics, MEMSM 2013
MEMS and Mechanics
, 10.4028/www.scientific.net/AMR.705.523, 523-527, 2013.07, [URL], This paper introduces a modified differential evoluiton method to solve the tension/compression string design problem. The modification is derived from mechanisms of social networks. In the proposed method, each individual will be attracted by the knowed best individual following the connectivity between each other. The connectivity is calculated based on the difference of the variables in each vector. The individuals with high connectivity tend to perform local search while those with poor connectivity tend to perform global search instead. The approach was employed for a tension/compression string design problem and by comparisons with the other evolutionary algorithms, the proposed method privided better resutls..
188. Makoto Okada, Sachio Hirokawa, Kiyota Hashimoto, A Investigation of Oppinions and Demands of New Farming Applicants of Japan Using Concept Graph in The Questionnaire Survey, Pcoc. IEIS2013, 104-104, 2013.06.
189. Makoto Okada, Sachio Hirokawa, Kiyota Hashimoto, Analysis of Opinions from Questionnaire Surveys of Farming Candidates using Cross, Proc. KES-IIMS2013, 184-194, 2013.06.
190. Xiao Lin, Eisuke Ito, Sachio Hirokawa, Chinese Tag analysis for foreign movie contents, Proc. of IEEE/ACIS ICIS2014, 163-166, 2013.06.
191. Bren Flanagan, Chengjiu Yin, Takahiko Suzuki, Sachio Hirokawa, Intelligent Computer Classification of English Writing Errors, Proc. KES-IIMS2013, 174-183, 2013.06, An important issue in education systems is the ability to determine
the characteristics of learners and then provide intelligent and
informed guidance in response. The authors of this paper have a
long-term research goal to provide language learners with the ability
to determine and improving their weaknesses. However, to achieve this
goal a sizable amount of manually classified data is required. The
task is both time consuming and labor intensive. In this paper a
system was built to help intelligently classify the errors in an
English learner’s writings into categories (Kroll 1990, Weltig
2004). Using a randomly selected manually classified sample as
training data, it was determined that there is a positive correlation
between the number of samples for each error category and the
effectiveness of the model created by applying SVM machine learning to
the writings of language learners on the Lang-8 website. It is
intended that the classification results will be used to accelerate
the manually process classification and increase the amount of
training data available for use..
192. CHENGJIU YIN, Sachio Hirokawa, Jane Yin-Kim Yau, Tetsuya Nakatoh, Kiyota Hashimoto, Yoshiyuki Tabata, Analyzing Research Trends with Cross Tabulation Search Engine, International Journal of Distance Education Technologies, 11, 1, 31-44, 2013.01.
193. Jun Zeng, Toshihiko Sakai, Chengjiu Yin, Takahiko Suzuki, Sachio Hirokawa, Automatic Generation of Tourism Quiz using Blogs, the journal Artificial Life and Robotics, 17, 3, 412-416, 2013.01.
194. Brendan Flanagan, Takahiko Suzuki, Jun Zeng, CHENGJIU YIN, Toshihiko Sakai, Kiyota Hashimoto, Sachio Hirokawa, Focused Search using Community Search Logs, Proc. ICDIPC2013, 559-565, 2013.01, Search engines have become an increasingly important educational
tool. Finding information has become an easy exercise by just typing a
few keywords into a search engine text-box. However, some people may
not use a search engine effectively. Students may not know how to
choose proper keywords. They may have difficulty in selecting the
relevant information from millions of search results. In this paper,
we propose a search engine system which shares search queries and
browsing history among students. We call the search queries and
browsing history “community search logs”. This system analyzes the
community search logs, and shares students’ knowledge and experience
to each other. Our purpose is with help students, especially those who
are not good at searching, to improve their searching efficiency..
195. Jun Zeng, Brendan Flanagan, Sachio Hirokawa, Layout-Tree-based Approach for Identifying Visually Similar Blocks in a Web Page, Proc. IWEA2013, 65-70, 2013.01, Search engines have become an increasingly important educational
tool. Finding information has become an easy exercise by just typing a
few keywords into a search engine text-box. However, some people may
not use a search engine effectively. Students may not know how to
choose proper keywords. They may have difficulty in selecting the
relevant information from millions of search results. In this paper,
we propose a search engine system which shares search queries and
browsing history among students. We call the search queries and
browsing history “community search logs”. This system analyzes the
community search logs, and shares students’ knowledge and experience
to each other. Our purpose is with help students, especially those who
are not good at searching, to improve their searching efficiency..
196. L. Jian, CHENGJIU YIN, Sachio Hirokawa, Yoshiyuki Tabata, Modified differential evolution for tension/Compression string design problem, Advanced Materials Research, 705, 523-527, 2013.01.
197. Brendan Flanagan, Takahiko Suzuki, Jun Zeng, CHENGJIU YIN, Toshihiko Sakai, Sachio Hirokawa, Kiyota Hashimoto, Peer Knowledge Assisted Search Using Community Search Logs, International Journal of Ditigal Information and Wireless Communications, 3, 2, 1-8, 2013.01.
198. Kazunori Shimizu, Eisuke Ito, Sachio Hirokawa, Predicting Future Ranking of Online Novels based on Collective Intelligence, Proc. ICDIPC2013, 263-274, 2013.01, A large number of novels are being uploaded as online novels. The
present paper proposes a ranking algorithm based on the users'
favorite lists (bookmarks). Empirical evaluation has been conducted
with respect to each genre of novels. In several genres, it is
confirmed that the top ranked novels in July are predicted from the
bookmarks of May.
.
199. Brendan Flanagan, Chengjiu Yin, Sachio Hirokawa, Kiyota Hashimoto, Yoshiyuki Tabata, An automated method to generate e-learning quizzes from online language learner writing, International Journal of Distance Education Technologies, 10.4018/ijdet.2013100105, 11, 4, 63-80, 2013.01, [URL], In this paper, the entries of Lang-8, which is a Social Networking Site (SNS) site for learning and practicing foreign languages, were analyzed and found to contain similar rates of errors for most error categories reported in previous research. These similarly rated errors were then processed using an algorithm to determine corrections suggested by a native speaker. Subject matter experts then evaluated the processed sentences to determine the quality in relation to use in tests or exams for language learners. The method describes the automatic generation of multiple choice and fll-in-the-blanks quizzes using the writings of language learners on public web based learning sites, in order to support learner refection on corrections and practicing past errors to overcome problems..
200. Jun Zeng, Toshihiko Sakai, Chengjiu Yin, Takahiko Suzuki, Sachio Hirokawa, Automatic generation of tourism quiz using blogs, Artificial Life and Robotics, 10.1007/s10015-012-0076-7, 17, 3-4, 412-416, 2013.01, [URL], The one-way information communication provision does not impress listeners well. As an efficient way, Question & Answer can not only help listeners to comprehend the content of information but also help information providers understand the response of listeners. However, it is not easy for everyone to create suitable questions. In this paper, we propose an automatic quiz generation system using tourism blogs. The system can generate quizzes by extracting feature words of the topic keyword from blogs. Our purpose is to help tourism information providers to advertise their tourism events in an interactive way, in order to impress tourists. When compared with other methods of quiz generation, we demonstrate that our method is more suitable for information communication provision..
201. Brendan Flanagan, Takahiko Suzuki, Jun Zeng, Chengjiu Yin, Toshihiko Sakai, Kiyota Hashimoto, Sachio Hirokawa, Focused search using community search logs, 3rd International Conference on Digital Information Processing and Communications, ICDIPC 2013
3rd International Conference on Digital Information Processing and Communications, ICDIPC 2013
, 557-563, 2013.01, Search engines have become an increasingly important educational tool. Finding information has become an easy exercise by just typing a few keywords into a search engine text-box. However, some people may not use a search engine effectively. Students may not know how to choose proper keywords. They may have difficulty in selecting the relevant information from millions of search results. In this paper, we propose a search engine system which shares search queries and browsing history among students. We call the search queries and browsing history "community search logs". This system analyzes the community search logs, and shares students' knowledge and experience to each other. Our purpose is with help students, especially those who are not good at searching, to improve their searching efficiency..
202. Chengjiu Yin, Sachio Hirokawa, Jane Yin Kim Yau, Kiyota Hashimoto, Yoshiyuki Tabata, Tetsuya Nakatoh, Research trends with cross tabulation search engine, International Journal of Distance Education Technologies, 10.4018/jdet.2013010103, 11, 1, 31-44, 2013.01, [URL], To help researchers in building a knowledge foundation of their research fields which could be a timeconsuming process, the authors have developed a Cross Tabulation Search Engine (CTSE). Its purpose is to assist researchers in 1) conducting research surveys, 2) efficiently and effectively retrieving information (such as important researchers, research groups, keywords), and also 3) providing analytical information relating to past and current research trends in a particular field. Their CTSE system employs data-processing technologies and emphasizes the use of a "Learn by Searching" learning strategy to support students to analyze such research trends. To show the effectiveness of CTSE, a pilot experiment has been conducted, where participants were assigned to do research survey tasks and then answer a questionnaire regarding the effectiveness and usability of the system. The results showed that the system has been helpful to students in conducting research surveys, and the research trend transitions that our system presented were effective for producing research trend surveys. Moreover, the results showed that most students had favorable attitudes toward the usage and usability of the system, and those students were satisfied in gaining more know ledge in a particular research field in a short period..
203. Makoto Okada, Sachio Hirokawa, Kiyota Hashimoto, A investigation of opinions and demands of new farming applicants of Japan using concept graph in the questionnaire survey, 2013 IEEE/ACIS 12th International Conference on Computer and Information Science, ICIS 2013
2013 IEEE/ACIS 12th International Conference on Computer and Information Science, ICIS 2013 - Proceedings
, 10.1109/ICIS.2013.6607824, 101-104, 2013, [URL], Agriculture is important field in Japan, but number of new farmer decreases or is flat in recent years. It is necessary for the agricultural field in Japan to increase new farmers. To solve the problem, it is important to investigate opinions and demands of new agricultural applicants. We investigated attitudes and interests in results of questionnaire surveys for the applicants. And we also investigate effectiveness of a method to extract and visualize these kinds of information contained in the questionnaire using concept graph..
204. Eisuke Ito, Brendan Flanagan, Chengjiu Yin, Tetsuya Nakatoh, Sachio Hirokawa, A private cloud environment for teaching search engine construction, 21st International Conference on Computers in Education, ICCE 2013
Proceedings of the 21st International Conference on Computers in Education, ICCE 2013
, 397-403, 2013, Kyushu University installed a private cloud system, named "campus cloud system", using VCL and CloudStack. For a graduate school exercise course on web search engine, the authors prepared a virtual machine on VCL, which had apache web server and GETA indexer preinstalled. This paper introduces an outline of the cloud system, the exercise, and also reports advantages and disadvantages of cloud based education..
205. Chengjiu Yin, Jane Yin Kim Yau, Sachio Hirokawa, Yoshiyuki Tabata, An SNS-based literature review system for conducting a research survey, 21st International Conference on Computers in Education, ICCE 2013
Proceedings of the 21st International Conference on Computers in Education, ICCE 2013
, 404-410, 2013, It is necessary to perform a literature review before starting a new research project. However, many students do not know the procedures of performing a literature review. In this paper, based on the professional experiences and opinions of expert researchers, we describe an SNS-based literature review system to help students conduct research surveys. This system includes two search engines, one is an article search engine, which can help students conduct research surveys, and the other is a logging search engine, which allows students to learn from each other via their logs and share experience with other students. User models of the system as well as its functions are presented..
206. Chengjiu Yin, Brendan Flanagan, Sachio Hirokawa, Yoshiyuki Tabata, Build a search engine to support doing research surveys on SNS, 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
Proceedings - 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
, 10.1109/IIAI-AAI.2013.11, 183-186, 2013, [URL], It is very important for any academic researches to do research surveys. In this paper, we proposed an SNS based search engine to support doing research surveys, called SNSearch. Our SNSearch system not only supports analysis of research trends, but also offers learners opportunities to reflect on their search behaviors. By browsing the experts' survey history, students can learn search skills..
207. Tetsuya Nakatoh, Sachio Hirokawa, Evaluation of tourism resources extraction based on Japanese dependency analysis, 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
Proceedings - 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
, 10.1109/IIAI-AAI.2013.77, 100-103, 2013, [URL], Blog articles by tourists contain interesting and personal experiences of where and how they have gone, what they have done and what they thought. Such individual experiences are helpful in many cases compared to the general and official information about the tourist resort by tourist agents. However, it is not easy to choose related articles and to extract still more nearly required information from these unsorted blog articles. We have proposed a method of feature extraction by dependency analysis of those sentences that describe tourist behavior. This paper apply the proposed method to 7,917,385 blog articles on Kyushu area and shows the evaluation about the obtained resources for tourism..
208. Brendan Flanagan, Chengjiu Yin, Takahiko Suzuki, Sachio Hirokawa, Intelligent computer classification of english writing errors, Intelligent Interactive Multimedia Systems and Services. Proceedings of the 6th International Conference on Intelligent Interactive Multimedia Systems and Services (IIMSS2013), 10.3233/978-1-61499-262-2-174, 254, 174-183, 2013, [URL], An important issue in education systems is the ability to determine the characteristics of learners and then provide intelligent and informed guidance in response. The authors of this paper have a long-term research goal to provide language learners with the ability to determine and improving their weaknesses. However, to achieve this goal a sizable amount of manually classified data is required. The task is both time consuming and labor intensive. In this paper a system was built to help intelligently classify the errors in an English learner's writings into categories (Kroll 1990, Weltig 2004). Using a randomly selected manually classified sample as training data, it was determined that there is a positive correlation between the number of samples for each error category and the effectiveness of the model created by applying SVM machine learning to the writings of language learners on the Lang-8 website. It is intended that the classification results will be used to accelerate the manually process classification and increase the amount of training data available for use..
209. Jun Zeng, Brendan Flanagan, Sachio Hirokawa, Layout-tree-based approach for identifying visually similar blocks in a web page, 2013 IEEE/ACIS 12th International Conference on Computer and Information Science, ICIS 2013
2013 IEEE/ACIS 12th International Conference on Computer and Information Science, ICIS 2013 - Proceedings
, 10.1109/ICIS.2013.6607818, 65-70, 2013, [URL], When extracting information from a web page, IE systems usually need to perform pattern recognition to identify the elements that have similar patterns. However, most of them are mainly based on analyzing HMTL source code, DOM tree, tag tree or Xpath of web pages. These methods are language-dependent, or more precisely, HTML-dependent. They have some insuperable limitations. In order to overcome these limitations, we propose a notion of layout-tree and a pattern recognition method to identify visual blocks with similar visual pattern using layout tree. In this paper, we call a visible rectangular region in a web page a visual block or block for short. We consider if the elements of two blocks are displayed in a similar layout, we define that the two blocks are visually similar. We first transform the layout into a layout tree. By calculating the similarity of the layout trees of two blocks, we can determine whether the two blocks are visually similar or not. The result of experiment shows that the layout tree is an effective method to identify visually similar blocks..
210. Chengjiu Yin, Han Yu Sung, Gwo Jen Hwang, Sachio Hirokawa, Hui Chun Chu, Brendan Flanagan, Yoshiyuki Tabata, Learning by searching
A learning environment that provides searching and analysis facilities for supporting trend analysis activities, Educational Technology and Society, 16, 3, 286-300, 2013, With the popularity of the Internet, online searching is becoming an important part of learning. In this paper, based on the "Learning by Searching" theory, a learning environment is developed, which includes a search engine to assist students in recognizing the progression of trends and keyword transitions for specific domains. To efficiently support research trend surveys, an automatic data accumulation and classification approach is proposed to construct the database excerpts instead of manual keyword registration or any other heuristic preprocesses. With an associative search module, the search engine dynamically searches for relevant words that are frequently used in the targeted academic field, and provides learners with effective visualizations to understand the trend transitions. An experiment has been conducted on a college information management course to show the effectiveness of the proposed approach. The experiment results show that the students who learned with the new approach had significantly better learning performance in terms of recognizing the trend transitions of the targeted issues than those who learned with conventional search engines..
211. Sachio Hirokawa, Message from the conference general chair, Quaternary International, 10.1109/IIAI-AAI.2013.4, 2013, [URL].
212. Kazunori Shimizu, Eisuke Ito, Sachio Hirokawa, Predicting future ranking of online novels based on collective intelligence, 3rd International Conference on Digital Information Processing and Communications, ICDIPC 2013
3rd International Conference on Digital Information Processing and Communications, ICDIPC 2013
, 261-272, 2013, A large number of novels are being upload-ed as online novels. The present paper pro-poses a ranking algorithm based on the us-ers' favorite lists (bookmarks). Empirical evaluation has been conducted with respect to each genre of novels. In several genres, it is confirmed that the top ranked novels in July are predicted from the bookmarks of May..
213. Tetsuya Nakatoh, Hirofumi Amano, Sachio Hirokawa, Prediction of growth rate of operating income using securities reports, 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
Proceedings - 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
, 10.1109/IIAI-AAI.2013.50, 84-88, 2013, [URL], Corporate analysis is needed for various purposes such as finding a good business partner or a good employment, as well as choosing a good investment. Conventionally, it has been based mainly on financial figures. Recent advances in natural language processing technology, however, has activated studies on analysis of non-financial, textual data. This paper tries to predict the growth rate of the operating income of a company from text data contained in the security report of that company. It reports that this method can classify profitable companies and loss-making ones at 55% F-measure..
214. Jun Zeng, Brendan Flanagan, Qingyu Xiong, Junhao Wen, Sachio Hirokawa, Proposal of seam degree and content similarity for web page segmentation, 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
Proceedings - 2nd IIAI International Conference on Advanced Applied Informatics, IIAI-AAI 2013
, 10.1109/IIAI-AAI.2013.56, 9-14, 2013, [URL], Page segmentation has received great attention in recent years. However, most research has been based on some pre-defined heuristics or visual cues which may be not suitable for large-scale page segmentation. In this paper, we proposed two parameters: seam degree and content similarity, to indicate the coherent degree of a page block. Instead of analyzing pre-defined heuristics or visual cues, our method utilizes the visual and content features to determine whether a page block should be divided into smaller blocks. We also proposed a principled page segmentation method using these two parameters. An experiment was conducted to determine the relationship between the two parameters and the number of segment results. The empirical results also show that our segmentation method can effectively segment a page into different semantic parts..
215. Hongjie Zhai, Makoto Haraguchi, Yoshiaki Okubo, Kiyota Hashimoto, Sachio Hirokawa, Shifting concepts to their associative concepts via bridges, 9th International Conference on International Conference on Machine Learning and Data Mining, MLDM 2013
Machine Learning and Data Mining in Pattern Recognition - 9th International Conference, MLDM 2013, Proceedings
, 10.1007/978-3-642-39712-7_45, 7988 LNAI, 586-600, 2013, [URL], This paper presents a pair of formal concept search procedures to find associative connection of concepts via bridge concepts. A bridge is a generalization of a sub-concept of an initial concept. The initial concept is then shifted to other target concepts which are conditionally similar to the initial one within the extent of bridge. A procedure for mining target concepts under the conditional similarity with respect to the bridge is presented based on an object-feature incident relation. Such a bridge concept is constructed in the concept lattice of person-feature incident relation. The latter incident relation is defined by aggregating the former document-feature relation to have more condensed relation, while keeping the variation of possible candidate bridges. Some heuristic rule, named Mediator Heuristics, is furthermore introduced to reflect user's interests and intention. The pair of these two procedures provides an efficient method for shifting initial concepts to target ones via some bridges. We show their usefulness by applying them to Twitter data..
216. Youhei Inokuchi, CHENGJIU YIN, Sachio Hirokawa, Bridging SNS ID and User Using NFC and SNS, Proc. ASID2012 (Anti-Counterfeiting, Security and Identification), 1-5, 2012.12.
217. Sachio Hirokawa, Makoto Okada, Kiyota Hashimoto, Extraction of Hints and Advice from Hotel Reviews for Improving Small Hotel Management, Proc. CEC(IEEE 14th International Conference on Commerce and Enterprise Computing), 166-170, 2012.12.
218. Toshihiko Sakai, Sachio Hirokawa, Feature Words that Classify Problem Sentence in Scientific Article, Proc. iiWAS2012, 360-367, 2012.12, Literature review requires understanding the contents from
several view points, such as the problem and the method
that the articles describe. Search from these viewpoints will
improve the e?ciency of survey, if particular segments of
articles were extracted, indexed and can be used as auxiliary
query. This paper focuses on sentences that describe the
problem in an abstract and the feature sets that classify such
problem sentences. Classi?cation performance are evaluated
by 10-fold cross-validation for six candidate sets of feature
words. It turned out that the set of all words gains the best
performance if 90% of the data are used as training data.
However, the set of a small number of words with positive
scores outperforms other feature sets, if the training data is
only 10%. In such a realistic situation, the feature words are
effective in improving classification performance..
219. Youhei Inokuchi, CHENGJIU YIN, Sachio Hirokawa, Generation of Hen-ai Map from Search Log for Foreign Language Learning, Proc. ACIS2012, The First Asian Conference on Information Systems, 115-118, 2012.12.
220. Sachio Hirokawa, Toward Multi-Lingual Knowledge Extraction from Travelers' Reviews, Proc. ACIS2012, The First Asian Conference on Information Systems, 119-122, 2012.12, Online customer reviews have been variously employed for text mining
and information retrieval in general. However, the result of those
analyses has to be well visualized for prospective innovations of
firms and enterprises that cannot afford a dedicated expert. In this
study, we collected thousands of online customer reviews of hotels on
TripAdvisor.com, and, among them, we analyzed those reviews written in
Chinese and Japanese to extract culture-related preferences. Our
analysis is based on a distance-biased mi-scores and provides a
visualization for interpretation. We interpreted the results and
confirmed our interpretation with a simple questionnaire
survey. Though it is quite primitive and extensive refinements are to
be required, this paper shows a good possibility to extract
culture-related preferences that will be useful for improvements and
innovations in tourism industries.
.
221. Tetsuya Nakatoh, Sachio Hirokawa, Extraction of tourist behavior contexts from blog by verbs and their objects, 1st IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
Proceedings of the 2012 IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
, 10.1109/IIAI-AAI.2012.31, 112-116, 2012.12, [URL], Blog articles by tourists contain interesting and personal experiences of where and how they have gone, what they have done and what they thought. Such individual experiences are helpful in many cases compared to the general and official information about the tourist resort by tourist agents. However, it is not easy to choose related articles and to extract still more nearly required information from these unsorted blog articles. This paper proposes a technique of feature extraction by dependency analysis of verbs and objects in those sentences that describe tourist's behavior. This paper applied the method to 7,917,385 blog articles on Kyushu area and reports some analysis on "where and what did they eat" as case studies..
222. Eisuke Ito, Sachio Hirokawa, Kazunori Shimizu, Introducing faceted views in diversity of online novels, 7th International Conference on Digital Information Management, ICDIM 2012
7th International Conference on Digital Information Management, ICDIM 2012
, 10.1109/ICDIM.2012.6360114, 2012.12, [URL], In recent years, user generated content services have become popular. The authors are interested in online novel services. Classification of online novels is difficult because keywords and genre are assigned by the author of the novel without control. In order to overcome the problem faced when category classifying and searching online novels, faceted views were introduced and a cross tabulation search and analysis system was developed. This system can discover relations between novel genres and keywords, and can find the author's trend..
223. Toshihiko Sakai, Brendan Flanagan, Jun Zeng, Tetsuya Nakatoh, Sachio Hirokawa, Search engine focused on multiple features of scientific articles, 1st IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
Proceedings of the 2012 IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
, 10.1109/IIAI-AAI.2012.51, 214-217, 2012.12, [URL], When starting new research or summarizing the results of research, it is necessary to review related work in the same research field. The research review requires several point of views such as "problem", "method", "result". Simple search by keywords is not effective to specify and narrow-down the scope of search for these meta purpose. In this paper, we focus on sentences to improve the efficiency of research survey with multiple viewpoints. We developed a search engine for scientific articles which classifies sentences by multiple viewpoints..
224. Tetsuya Nakatoh, Sachio Hirokawa, Extraction of Tourist Behavior Contexts from Blog by Verbs and Their Objects, Proc. ESKM 2012, 112-116, 2012.09, Blog articles by tourists contain interesting and
personal experiences of where and how they have gone, what they
have done and what they thought. Such individual experiences
are helpful in many cases compared to the general and official
information about the tourist resort by tourist agents. However,
it is not easy to choose related articles and to extract still
more nearly required information from these unsorted blog
articles. This paper proposes a technique of feature extraction
by dependency analysis of verbs and objects in those sentences
that describe tourist’s behavior. This paper applied the method
to 70,352 blog articles on Kyushu area and reports some analysis
on "where and what did they eat" as case studies.
.
225. Chengjiu Yin, Sachio Hirokawa, Brendan Flanagan, Takahiko Suzuki, Yoshiyuki Tabata, Mistake discovery and generation of exercises automaticity in contex, Proc. LTLE2012, 163-167, 2012.09, It is useful for learners to write in a foreign language and ask
another person to correct it for them so they can know their
mistakes. However, a problem for learners' is how to reflect on the
correction of their problems and then master the mistake. In this
paper, by analyzing the entries of Lang-8, which is a SNS site for
learning and practicing foreign languages, we found the correspondence
between original entries and the corrections. Based on this finding, a
system was developed which can generate fill-in-the-blanks quizzes
according to different situations, in order to support learner's
reflection on correction and practicing so the mistake is no longer a
problem.
.
226. Hiroaki Ninomiya, Brendan Flanagan, Eisuke Ito, Sachio Hirokawa, Near friends communication encouragement system using NFC and SNS, Proc. ESKM2012, 20-22, 2012.09.
227. Toshihiko Sakai, Kiyota Hashimoto, Yuta Kamisoyama, Makoto Okada, Sachio Hirokawa, Polarity Estimation of Tweets by Feature Sets, Proc. CAINE2012, 105-110, 2012.09.
228. Toshihiko Sakai, Brendan Flanagan, Jun Zeng, Tetsuya Nakatoh, Sachio Hirokawa, Search Engine Focused on Multiple Features of Scientific Articles, Proc. ESKM2012, 214-217, 2012.09.
229. Shin-ichiro Yoshida, Tetsuya Nakatoh, Shuichi Mitarai, Sachio Hirokawa, Text Mining of Securities Reports for Discoverng Reason of Change, Proc. CAINE2012, 41-45, 2012.09, A stock market is a base of the economic activity
in present-day free economy society. In order to become
a listed company, there is a severe examination.
Furthermore, yearly duty is attached by law for the
listed company to file the fixed form report about the
financial condition. In this paper, correspondence of
the numeric data about the activity of a company and
text data is analyzed for the financial report. All of 68
medical-supplies-related listed companies were chosen
first. Then, the feature words were extracted from the
reports of the companies where achievements have improved
favorably. Moreover, the feature words of the
financial report of the time were extracted about the
companies where achievements carried out the sharp
reversal to profitability..
230. Eisuke Ito, Kazunori Shimizu, Sachio Hirokawa, Introducing faceted views in diversity of online novels, Proc. International Conference on Digital Information Management (ICDIM2012), 2012.08.
231. T. Sakai, M. Matsushita, B. Flanagan, J. Zeng, S. Hirokawa, Analysis of Influence of Investor Relation Documents to Stock Price, Proc. FSKD2012, 2012.05.
232. Toshihiko Sakai,Jun Zeng,Brendan Flanagan,Tetsuya Nakatoh,Sachio Hirokawa, Discriminant Words for Problems in Scientific Articles, Proceedings of International Symposium on Innovative E-Services and
Information Systems (IEIS 2012)
, 2012.05.
233. J. Zeng,T. Sakai, B. Flanagan, S. Hirokawa, Extraction of Relevant Components Using Shallow Structure of HTML Documents, Proc. FSKD2012, 2012.05.
234. Sachio Hirokawa, Feature Extraction using Restricted Bootstrapping, Proceedings of International Symposium on Innovative E-Services and
Information Systems (IEIS 2012)
, 2012.05.
235. Brendan Flanagan, Chengjiu Yin, Sachio Hirokawa, Han-Yu Sung and Gwo-Jen Hwang, Analysing Research Trends of Mobile Learning with the Milky Way, Proc. 7th International Conference on Wireless, Mobile and Ubiquitous Technology in Education (WMUTE2012), 249-253, 2012.03.
236. Tetsuya Nakatoh, Chengjiu Yin and Sachio Hirokawa, Characteristic Grammatical Context of Tourism Information, ICIC Express Letters, 6, 2, 563-568, 2012.03.
237. Jun Zeng, Sachio Hirokawa, Component Search Engine based on HTML Path and Word Weight, ICIC Express Letters, 6, 3, 753-758, 2012.03.
238. Tetsuya Nakatoh, Chengjiu Yin, Sachio Hirokawa, Characteristic grammatical context of tourism information, ICIC Express Letters, 6, 3, 753-758, 2012.03, We aim at constructing the ontology of tourism by gathering the resources of tourism on WWW. For the purpose, we paid attention to actual behavior of tourists performed at each tourist resort described in blog articles. Gathering actual behavior of tourists appropriately enables extraction of a set of typical behavior of tourists by a statistical method. This paper reports the attempt of basic extraction of tourist behavior by Japanese dependency analysis..
239. Jun Zeng, Toshihiko Sakai, Brendan Flanagan, Sachio Hirokawa, Extraction of Feature Words with the Same Generality Level as Query using Restricted Bootstrapping, Proc. 11th International Conference on Computer and Information Science(ICIS), 283-288, 2012.02.
240. Jun Zeng, Sachio Hirokawa, Component search engine based on html path and word weight, ICIC Express Letters, 6, 2, 563-568, 2012.02, With the popularization of search engines, finding information has become easier than before. However, the information found by most search engines today is web page, which may contain more than one topic. It makes user spend extra time to read the irrelevant contents in order to find out the information he wants. We propose a novel search engine model called "Component Search Engine", which calculates the score of each component in a page by HTML path and word weight. The higher score the component gains, the higherranking it will appear at. The usability study determinates that the component search engine can find out the important contents efficiently..
241. Jun Zeng, Toshihiko Sakai, Chengjiu Yin, Takahiko Suzuki, Sachio Hirokawa, Automatic Generation of Tourism Quiz using Blogs, 7th International Symposium on Artificial Life and Robotics(AROB2012), 97-100, 2012.01, There are large amount of Blog articles concerning to tourism.
There are many interests in searching and reading those articles.
This paper proposes a method to convert the articles as amusement
media where users can enjoy answering and learning well-known fact
or unexpected linkage.
242. Xiaobin Wu, Zeng Jun, Chengjiu Yin, Sachio Hirokawa, Sharing Knowledge and Experience of Search with SNS, 7th International Symposium on Artificial Life and Robotics(AROB2012), 101-104, 2012.01, Search Engine is an essential tool when we search and learn something
we do not know. We can share the search result if the answer were simple.
However, sharing the awareness and the process of search is hard to share
or tell other people. This paper proposes a community portal that
combines SNS and search engine.
243. Kiyota Hashimoto, Kazuhiro Takeuchi, Makoto Okada, Sachio Hirokawa, Visual chance discovery method of potential keys for innovations in tourism, 7th International Symposium on Artificial Life and Robotics(AROB2012), 89-92, 2012.01.
244. Brendan Flanagan, Chengjiu Yin, Sachio Hirokawa, Han Yu Sung, Gwo Jen Hwang, Analysing research trends of mobile learning with the Milky Way, 2012 17th IEEE International Conference on Wireless, Mobile and Ubiquitous Technology in Education, WMUTE 2012
Proceedings 2012 17th IEEE International Conference on Wireless, Mobile and Ubiquitous Technology in Education, WMUTE 2012
, 10.1109/WMUTE.2012.61, 249-253, 2012, [URL], Identifying research trends is an integral part of the planning phase of academic research and can be a daunting task for those not familiar with the process. This paper proposes the use of the Milky Way search engine to assist novice students who are beginning to undertake research. Using data-processing and data-mining techniques with data from SciVerse Scopus, the Milky Way search engine can provide students with a way to investigate research trends and test trend hypothesis' in their field. We evaluated the use of the Milky Way search engine in the field of mobile learning to analyze research trends..
245. Hiroaki Ninomiya, Eisuke Ito, Brendan Flanagan, Sachio Hirokawa, Bridging SNS ID and user using NFC and SNS
Design of NFC and SNS based event attendance management system, 2012 International Conference on Anti-Counterfeiting, Security and Identification, ASID 2012
2012 International Conference on Anti-Counterfeiting, Security and Identification, ASID 2012
, 10.1109/ICASID.2012.6325346, 2012, [URL], A smart phone and a tablet terminal with a NFC function are spreading by standardization of NFC technology. The authors are interested in the exchange support between the participants in the off-line meeting. The authors focus on participants' mutual communication in off-line meetings. It is rare to perform exchange in an actual meeting place, although there is a relation called a friend or a follower on on-line. This will be because of gap between identity on SNS and identity in the real world. The authors are interested in bridging the gap between the online identity and the identity in the real world. The authors are developing a system, which matches human on online SNS ID and a real person, using a smart phone or a mobile terminal with NFC technology. Our system will display the human relations on the on-line SNS, and connect a meeting and a meeting participant..
246. Toshihiko Sakai, Jun Zeng, Brendan Flanagan, Tetsuya Nakatoh, Sachio Hirokawa, Descriminant words for problems in scientific articles, 2012 IEEE/ACIS 11th International Conference on Computer and Information Science, ICIS 2012
Proceedings - 2012 IEEE/ACIS 11th International Conference on Computer and Information Science, ICIS 2012
, 10.1109/ICIS.2012.42, 267-271, 2012, [URL], Various viewpoints are required to make a survey and a trend analysis on related research. In order to find important problems, especially in unfamiliar field, simple search and clustering is not enough. We have to read most of the articles carefully. The work requires a lot of time and effort. This paper analyzes the sentences that describe the problem using SVM. It turned out the negative words are more effective in discernment than manually selected clue words or the positive words..
247. Jun Zeng, Toshihiko Sakai, Brendan Flanagan, Sachio Hirokawa, Extraction of feature words with the same generality level as query using restricted bootstrapping, 2012 IEEE 14th International Conference on Commerce and Enterprise Computing, CEC 2012
Proceedings of the 2012 IEEE 14th International Conference on Commerce and Enterprise Computing, CEC 2012
, 10.1109/CEC.2012.40, 171-176, 2012, [URL], It is not so simple to get an appropriated level of search result among a large number of targets which might be too general or too specific. Hints for the query are valuable for a user to expand or shrink his next search step, if the hints would be shown with their levels compared with the user's original query. This paper proposes a method to extract feature words of the same level as user's query using restricted bootstrap. Examples are shown to demonstrate the effectiveness of the method on tourism blogs. The paper proposes an evaluation measure for the similarity of levels for words based on WordNet..
248. Sachio Hirokawa, Makoto Okada, Kiyota Hashimoto, Extraction of hints and advice from hotel reviews for improving small hotel management, 2012 IEEE 14th International Conference on Commerce and Enterprise Computing, CEC 2012
Proceedings of the 2012 IEEE 14th International Conference on Commerce and Enterprise Computing, CEC 2012
, 10.1109/CEC.2012.37, 166-170, 2012, [URL], There are various kinds of and huge amounts of hotel information both from providers and customers. Hotel information by hotels and travel agents are reliable. A much large number of reviews by general users are available, which might be less reliable compared to the official ones. However, those reviews are helpful, since we can hear their personal experience and opinion. This paper proposes a method to find hints and advice for improving small hotel management by extracting and analyzing the feature words of reviews. This paper focuses on the secondary major words and compares their occurrence probabilities in the business customer's review with the family customer's review. A visual interpretation is proposed by mapping the feature words on Word Net..
249. Jun Zeng, Brendan Flanagan, Toshihiko Sakai, Sachio Hirokawa, Extraction of relevant components using shallow structure of HTML documents, 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2012
Proceedings - 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2012
, 10.1109/FSKD.2012.6234295, 1186-1190, 2012, [URL], As the amount of web page increases, searching for semi-structured documents is gaining greater attention. The traditional approach for extracting data from web page documents is to write specialized programs, called wrappers that identify data of interest and map them to some suitable format. However, developing wrappers manually has many well known shortcomings, mainly due to the difficulty in writing and maintaining them for continually changing web data. Moreover, there is no one wrapper program that can treat all kinds of web pages. In this paper, we aim to extract relevant and meaningful snippets from as many web pages as possible, using the shallow feature of HTML documents to discover and analyze the relevant components. Also, we introduced a new feature called GAP and verified the effectiveness of GAP by conducting a SVM learning experiment..
250. Jun Zeng, Junhao Wen, Qingyu Xiong, Sachio Hirokawa, Extraction of relevant snippets from web pages using hybrid features, 1st IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
Proceedings of the 2012 IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
, 10.1109/IIAI-AAI.2012.50, 209-213, 2012, [URL], As the amount of web pages increase, identifying and retrieving distinct contents from the web has increasingly become more and more difficult. The traditional approach for extracting data from web page documents is to analyze the DOM (Document Object Model) structure of a HTML page and find a common pattern. However, the number of possible DOM layout patterns is virtually infinite, which means that there is no common pattern that can be used for all kinds of web pages. In this paper, we focus on the pages that are linked to a search engine and aim to analyze the features of relevant and meaningful contents instead of a common pattern. Three features of relevant snippets are introduced. They are: quantity of text, correlation between snippet and query that is inputted into a search engine, and HTML structure. Nine parameters are used to describe the three features. Also, a SVM learning experiment is conducted to verify the effectiveness of the three features. The results show that the HTML structure feature is the most effective feature which can determine whether a snippet is relevant or not..
251. Sachio Hirokawa, Feature extraction using restricted bootstrapping, 2012 IEEE/ACIS 11th International Conference on Computer and Information Science, ICIS 2012
Proceedings - 2012 IEEE/ACIS 11th International Conference on Computer and Information Science, ICIS 2012
, 10.1109/ICIS.2012.50, 283-288, 2012, [URL], The bootstrapping method is known as an application of the Page-rank technique for documents and words. The technique calculates the score of the words by mutually propagating the score of the words and the documents. However, sometimes the result is far away from the initial query word. The problem is known as "topic drift". This paper proposes to restrict the words to be to the top t words in the process of bootstrapping. The method is simpler than the technique known so far. The method is applied for the real bankruptcy information documents to extract the bankruptcy causes strongly related to the query. It is confirmed that the method prevents the topic drift..
252. Toshihiko Sakai, Sachio Hirokawa, Feature words that classify problem sentence in scientific article, 14th International Conference on Information Integration and Web-Based Applications and Services, iiWAS 2012
14th International Conference on Information Integration and Web-Based Applications and Services, iiWAS 2012 - Proceedings
, 10.1145/2428736.2428803, 360-367, 2012, [URL], Literature review requires understanding the contents from several view points, such as the problem and the method that the articles describe. Search from these viewpoints will improve the efficiency of survey, if particular segments of articles were extracted, indexed and can be used as auxiliary query. This paper focuses on sentences that describe the problem in an abstract and the feature sets that classify such problem sentences. Classification performance are evaluated by 10-fold cross-validation for six candidate sets of feature words. It turned out that the set of all words gains the best performance if 90% of the data are used as training data. However, the set of a small number of words with positive scores outperforms other feature sets, if the training data is only 10%. In such a realistic situation, the feature words are effective in improving classification performance..
253. Sachio Hirokawa, Message from the conference general chair, Quaternary International, 10.1109/IIAI-AAI.2012.5, 2012, [URL].
254. Chengjiu Yin, Sachio Hirokawa, Brendan Flanagan, Takahiko Suzuki, Yoshiyuki Tabata, Mistake discovery and generation of exercises automaticity in context, 1st IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
Proceedings of the 2012 IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
, 10.1109/IIAI-AAI.2012.41, 163-167, 2012, [URL], It is useful for learners to write in a foreign language and ask another person to correct it for them so they can know their mistakes. However, a problem for learners' is how to reflect on the correction of their problems and then master the mistake. In this paper, by analyzing the entries of Lang-8, which is a SNS site for learning and practicing foreign languages, we found the correspondence between original entries and the corrections. Based on this finding, a system was developed which can generate fill-in-the-blanks quizzes according to different situations, in order to support learner's reflection on correction and practicing so the mistake is no longer a problem..
255. Hiroaki Ninomiya, Eisuke Ito, Brendan Flanagan, Sachio Hirokawa, Near friends communication encouragement system using NFC and SNS, 1st IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
Proceedings of the 2012 IIAI International Conference on Advanced Applied Informatics, IIAIAAI 2012
, 10.1109/IIAI-AAI.2012.37, 145-148, 2012, [URL], The value of an open event is in active communication betweenparticipants. However, it is difficult to look for those who have a common interest or a common friend. You would miss an opportunity to speak directly to your SNS friends or to those who have common SNS friends, unless you notice that there are such participants and you recognize them. This paper proposes a participant managerial system for using the human relations in SNS in an actual event site. Each participant makes his registration to an event using his SNS ID. In the event site on the day, the participant only has to touch his mobile terminal to an NFC tag. The participant and his SNS are bound by this action. The organizer as well as the participants can understand the background and the interests of each participant using SNS information. NFC tag provides real time attendance-and-absence status of the participants. It is expectable that the proposed system promotes exchange between participants..
256. Toshihiko Sakai, Kiyota Hashimoto, Yuya Kamisoyama, Makoto Okada, Sachio Hirokawa, Polarity estimation of tweets by feature sets, 25th International Conference on Computer Applications in Industry and Engineering, CAINE 2012 and the 4th International Symposium on Sensor Network and Application, SNA 2012
25th International Conference on Computer Applications in Industry and Engineering, CAINE 2012 and 4th International Symposium on Sensor Network and Application, SNA 2012
, 105-110, 2012, Sentiment analysis of micro-blogs, such as twitter, is one of the hottest topics. The speed of information propagation is becoming faster and faster. We cannot control the flow of information on tweets. So, we need to know the characteristics of such communication tools. The present paper extracts the features of emotional tweets, based on feature selection by SVM. An attention is paid to part of speech, particularly to the particles..
257. Toshihiko Sakai, Masashi Matsushita, Brendan Flanagan, Jun Zeng, Sachio Hirokawa, RETRACTED ARTICLE
Analysis of influence of investor relation documents to stock price, 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2012
Proceedings - 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2012
, 10.1109/FSKD.2012.6234291, 1280-1284, 2012, [URL], Not only specialists but also ordinary people have interests in the stock price changes. The present paper analyzes the influence of the frequency of characteristic words that appear in the investor relation (IR) documents of companies. Is it true that the positive words cause the rise of a stock price, and the negative words cause the fall? We focused on 57 medical-supplies-related companies which have listed on Tokyo Stock Exchange. The stock price data and IR documents of the period in 2009 to 2011 were collected. We prepared 10 positive words and 10 negative words to evaluate the score of IR documents. The rate of the stock price compared with the previous day and the document score was analyzed. It is confirmed that the correlation of the rate and the document score is higher for the companies with high stock price ratio of change..
258. Shin Ichiro Yoshida, Tetsuya Nakatoh, Shuichi Mitarai, Sachio Hirokawa, Text mining of securities reports for discovering reason of change, 25th International Conference on Computer Applications in Industry and Engineering, CAINE 2012 and the 4th International Symposium on Sensor Network and Application, SNA 2012
25th International Conference on Computer Applications in Industry and Engineering, CAINE 2012 and 4th International Symposium on Sensor Network and Application, SNA 2012
, 41-45, 2012, A stock market is a base of the economic activity in present-day free economy society. In order to become a listed company, there is a severe examination. Furthermore, yearly duty is attached by law for the listed company to file the fixed form report about the financial condition. In this paper, correspondence of the numeric data about the activity of a company and text data is analyzed for the financial report. All of 68 medical-supplies-related listed companies were chosen first. Then, the feature words were extracted from the reports of the companies where achievements have improved favorably. Moreover, the feature words of the financial report of the time were extracted about the companies where achievements carried out the sharp reversal to profitability..
259. Chengjiu Yin, Yoshiyuki Tabata, Sachio Hirokawa, A Milky Way Research Trend system for Survey of Scientific Literature, Proc. ELSM 2011, First International Workshop on Enhancing Learning with Social Media, 2011.12.
260. Chengjiu YIN, Yoshiyuki TABATA, Kiyota HASHIMOTO, Tetsuya NAKATOH, Sachio HIROKAWA, A Support System for Research Trend Survey of Scientific Literature, Proc. ICCE2011, 116-118, 2011.12.
261. Sachio Hirokawa, Jun Zeng, Comparison of Tourism Data using Double Ranking, Proc. SSNE, International Workshop on Innovative Tourism Informatics, 67-72, 2011.12, The prompt discovery and mining of user’s reputation, opinion and complaint are becoming a hot topic in the field of search engine for Blog and Twitter. This paper proposes "Double Rank" method to analyze the search result using two viewpoints, where the polarity degree of keywords in the search results are evaluated from the viewpoints. Most previous researches concern mainly in positive and negative evaluation as for the polarity degree. Two viewpoints are specified as the search condition in the present paper. The blog articles related to sightseeing are analyzed as case studies, where the characteristics and the sightseeing situation are compared for two prefectures..
262. Tetsuya Nakatoh, Chengjiu Yin, Sachio Hirokawa, Extraction and Disambiguation of Name of Place from Tourism Blogs, Proc. SSNE, International Workshop on Innovative Tourism Informatics, 73-78, 2011.12, By development of the Internet in recent years,
tourism portal sites and blog articles about tourism increased
on WWW. Acquisition of various tourism information became
easy. When gathering and classifying the information automatically
from blog articles, it is not easy to decide automatically
place names used as the key. In this paper, we propose a method
of extracting place names from blog articles automatically.
Moreover, we also tried disambiguation of a place name..
263. Chengjiu Yin, Eisuke Ito, Tetsuya Nakatoh, Sachio Hirokawa, Classification Network of tourism information
A smart phone-based system for supporting "Petit Trips", 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
Proceedings of 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
, 10.1109/ICAwST.2011.6163134, 170-173, 2011.12, [URL], We are developing a concept dictionary for Petit Trips (Short Time Trips). Based on the concept dictionary, we propose a smart phone-based system to support the Petit Trips. This system provides travelers a Classification Network of tourist spots according to their locations. While travelers select their preferred tourist spots, the system will recommend the suitable tourist spots which have similar features. Through the Classification Network of tourist spots, travelers can go to their preferred tourist spots one after another..
264. Tetsuya Nakatoh, Chengjiu Yin, Hiroki Matsuura, Sachio Hirokawa, Visualization of tourism information using WordNet, 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
Proceedings of 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
, 10.1109/ICAwST.2011.6163110, 412-417, 2011.12, [URL], In recent years, with the development of the Internet, and the rapidly increasing number of tourism portal sites and blogs, we can obtain a variety of tourist information on the Internet. If we have a specific need, we can obtain the required information through checking the retrieved results one by one. However, in order to see the whole trend of the results, it is necessary to analyze and visualize the document group. In this paper, by storing a set of tourism-related future words in WordNet, which is a concept thesaurus, we proposed a method using information from WordNet to visualize a document group as a conceptual graph. With the structure of a thesaurus, this method enables us to understand the contents of document group at a glance. Furthermore, by comparing the thesaurus structures, which are obtained from different background document groups, we can grasp the differences..
265. Kensuke Baba, Toshie Tanaka, Emi Ishita, Masao Mori, Eisuke Ito, Sachio Hirokawa, Evaluation of link system between repository and researcher database, 13th International Conference on Asia-Pacific Digital Libraries, ICADL 2011
Digital Libraries
For Cultural Heritage, Knowledge Dissemination, and Future Creation - 13th International Conference on Asia-Pacific Digital Libraries, ICADL 2011, Proceedings
, 10.1007/978-3-642-24826-9_50, 381-382, 2011.11, [URL], This paper evaluates the effect of a Web system which activates institutional repositories. Institutional repository is an important service of libraries in academic institutions. The authors developed a link system between the institutional repository and the researcher database of their university. The system reduces the efforts of researchers by reusing the metadata in the researcher database for registrations of their papers to the repository. The authors observed the access log of the repository before and after the start up of the link system. The result shows that the system increased the number of access, however there was no significant change on the number of registration of papers..
266. C. Yin, Y.Tabata, X. Wu, T. Nakatoh, S. Hirokawa, Building a Search Engine for Scientific Projects Survey, Proc. of IEEE Internetional Conference on Cyber, Physical and Social Computing
(CPSCom 2011) workshops on "Technology-Enhanced Social Learning"
, 558-563, 2011.10, This paper targets the students, who are just beginning to
engage in research. With the data-mining technologies, using the data
of KAKEN (Grant-in-Aid for Scientific Research of Japan), according to
students' learning styles and learners' knowledge levels, we propose
to create a "Learning by Searching" search engine to provide suitable
knowledge and help students to master research trends.
.
267. K. Baba, T. Tanaka, E. Ishita, M. Mori, E. Ito, S. Hirokawa, Evaluation of Link System between Repository and Researcher Database, Proc. ICADL 2011, LNCS 7008, 382--383, 2011.10.
268. Kiyota Hashimoto, Kazuhiro Takeuchi, Chengjiu Yin, Sachio Hirokawa, Extraction of subjective context-sensitive evaluation of Japanese onomatopoeic expressions and its applications, ICIC Express Letters, 5, 10, 3755-3760, 2011.10, Onomatopoeic expressions are frequently used in Japanese but, as well as other types of subjective adjectival and adverbial expressions, their Kansei evaluative value, or their Kansei polarity, depends on genre and context and is difficult to describe, let alone their exact meaning. In this study, an automatic extraction method of Kansei polarities of Japanese onomatopoeic expressions according to genre and context is proposed. It uses the contrastive use of pair expressions with seemingly opposite Kansei polarities in different genres and contexts, and the results of our experiments indicate that our method is effective for better descriptions of onomatopoeic expressions..
269. Sachio Hirokawa, Takahiro Baba, Tetsuya Nakatoh, Search and analysis of bankruptcy cause by classification network, 1st International Conference on Model and Data Engineering, MEDI 2011
Model and Data Engineering - First International Conference, MEDI 2011, Proceedings
, 10.1007/978-3-642-24443-8_17, 6918 LNCS, 152-161, 2011.10, [URL], A simple document search is insufficient when we analyse corporate information. Not only a list of search results, but also a justification why the results match the query condition is important. This paper proposes a method to extract cause of bankruptcy from news articles applying the co-occurrence analysis of words..
270. Yoshiaki Okubo, Makoto Haraguchi, Sachio Hirokawa, Finding Top-N Chance Patterns with KeyGraph-Based Importance, Proc. KES2011, LNCS 6882, 457--468, 2011.09.
271. Sachio Hirokawa, Takahiro Baba, Tetsuya Nakatoh, Search and Analysis of Bankruptcy Cause by Classification Network, Proc. 1st International Conference on Model & Data Engineering MEDI2011,
LNCS 6918
, 327--331, 2011.09.
272. Sachio Hirokawa, Chengjiu Yin, Tetsuya Nakatoh, Component-based search engine for blogs, 2011 IEEE International Conference on Fuzzy Systems, FUZZ 2011
FUZZ 2011 - 2011 IEEE International Conference on Fuzzy Systems - Proceedings
, 10.1109/FUZZY.2011.6007650, 1074-1078, 2011.09, [URL], A wrapper is a program that selectively extracts a necessary part (component) from Web pages. Automatic or semi-automatic wrapper construction is crucial to achieve a fine grained search engine for Web pages. However, this is not an easy task to achieve. This paper proposes a component-based search engine in which the content components gain a high score in the search results. Thus, the required segments for a query can be obtained without using a wrapper..
273. Yoshiaki Okubo, Makoto Haraguchi, Sachio Hirokawa, Finding top-N chance patterns with KeyGraph®-based importance, 15th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2011
Knowledge-Based and Intelligent Information and Engineering Systems - 15th International Conference, KES 2011, Proceedings
, 10.1007/978-3-642-23863-5_47, 457-468, 2011.09, [URL], In this paper, as our first proposal, we discuss a method for finding a rare pattern, called a chance pattern, which connects a pair of more frequent patterns. Particularly, our chance pattern is defined with a KeyGraph®-based importance of patterns. More concretely speaking, a chance pattern is a pattern C which often appears in a part of documents containing a frequent pattern XL as well as in those containing another pattern XR, that is, confidence values of association rules, XL ⇒ C and X R ⇒ C, are relatively high. It would be expected that such a chance pattern C reveals a hidden and implicit relationships between X L and XR. We design clique-search-based algorithms for finding chance patterns with Top-N confidence values..
274. Chengjiu Yin, Tetsuya Nakatoh, Ito Eisuke, Sachio Hirokawa, Classification network of tourism information -- A smart phone-based
system for supporting "Petit Trips", Proc. iCAST2011, 171--174, 2011.08.
275. J. Zeng, S. Hirokawa, Component Search Engine using Tree-View Interface, Proc. iCAST2011, 327--331, 2011.08.
276. S. Hirokawa, C. Yin, K. Hashimoto, Multidisciplinary Multi-Faceted Search Engine of Literatures on Tourism, Proc. 2nd International Symposium on Applied Informatics (ISAI 2011), 2011.08.
277. S. Hirokawa, T. Baba, T. Nakatoh, Text Mining of Bankruptcy Information using Formal Concept Analysis, Proc. iCAST2011, 532--537, 2011.08.
278. T. Nakatoh, C. Yin, H. Matsuura, S. Hirokawa, Visualization of Tourism Information using WordNet, Proc. iCAST2011, 413--418, 2011.08.
279. Sachio Hirokawa, Chengjiu Yin, Kiyota Hashimoto, Kazuhiro Takeuchi, Search and analysis of gourmet blogs with a particular reference to onomatopoeia, ICIC Express Letters, 5, 8 B, 2971-2976, 2011.08, Personal blogs contain many articles of the dish, sweets, impression and experience when the blogger visited the location. This kind of information is hardly obtained from pamphlets or homepages of hotels and travel agents. This paper proposes a search engine that focuses on the usage of onomatopoeic words that appear on these blogs..
280. Sachio Hirokawa, Chengjiu Yin, Tetsuya Nakatoh, Component-Based Search Engine for Blogs, Proc. FUZZ-IEEE, 2011.06.
281. Kensuke Baba, Masao Mori, Eisuke Ito, Sachio Hirokawa, A Feedback System on Institutional Repository, Proc. INTENSIVE (3rd International Conference on Resource Intensive Applications and Services), pp.37-42, 2011.05.
282. Masao Mori, Toshie Tanaka, Sachio Hirokawa, A Progressive Data Warehouse of Institutional Research with Web API and Mashup Visualization, Proc. CSEDU (3rd International Conference on Computer Supported Education), 2011.05.
283. Kiyota Hashimoto, Kazuhiro Takeuchi, Chengjiu Yin, Sachio Hirokawa, Extraction of Subjective Context-Sensitive Evaluation of Japanese Onomatopoeic Expressions and its Applications, ICIC Express Letters
, Vol.5, No.10, pp.3755-3760, 2011, 2011.05.
284. Sachio Hirokawa, Chengjiu Yin, Kiyota Hashimoto, Kazuhiro Takeuchi, Search and Analysis of Gourmet Blogs with a Particular Reference to Onomatopoeia, ICIC Express Letters
, Vol.5, No.8(B), pp.2971-2978, 2011, 2011.05.
285. Kensuke Baga, Eisuke Ito, Sachio Hirokawa, Co-occurrence Analysis of Access Log of Institutional Repository, Proc. JCAICT (Japan-Cambodia Joint Symposium on Information Systems and Communication Technology), pp.25--29, 2011, 2011.01.
286. Xiaobin Wu, Sachio Hirokawa, Chengjiu Yin, Tetsuya Nakatoh, Yoshiyuki Tabata, Extraction and Comparison of Tourism Information on the Web, Proc. AROB16 (16th International Symposium on Artificial Life and
Robotics)
, pp.228--231, 2011, 2011.01.
287. Masao Mori, Toshie Tanaka, Sachio Hirokawa, A progressive datawarehouse of institutional research with web API and mashup visualization, 3rd International Conference on Computer Supported Education, CSEDU 2011
CSEDU 2011 - Proceedings of the 3rd International Conference on Computer Supported Education
, 2, 323-329, 2011, We propose a progressive data warehouse which provides functions of operating statistics and their visualization for institutional research. The proposed data warehouse has a mashup programming environment with GUI and the users can share their programs. By sharing programs of data analysis, persons in charge of self-assessment seize an opportunity not only to create reports efficiently, but also to be able to improve their activities..
288. Chengjiu Yin, Yoshiyuki Tabata, Kiyota Hashimoto, Tetsuya Nakatoh, Sachio Hirokawa, A support system for research trend survey of scientific literature, 19th International Conference on Computers in Education, ICCE 2011
Proceedings of the 19th International Conference on Computers in Education, ICCE 2011
, 116-118, 2011, We constructed a support system for research trend surveys not only to accelerate the preliminary step but also to let students have a better grips of trend progresses and keyword transitions. Our system dynamically searches relevant words that are frequently used in the targeted academic field and gives users effective visualizations to understand trend transitions..
289. Chengjiu Yin, Yoshiyuki Tabata, Xiaobin Wu, Tetsuya Nakatoh, Sachio Hirokawa, Building a Search Engine for Scientific Projects Survey, 2011 IEEE International Conference on Internet of Things, iThings 2011 and 4th IEEE International Conference on Cyber, Physical and Social Computing, CPSCom 2011
Proceedings - 2011 IEEE International Conferences on Internet of Things and Cyber, Physical and Social Computing, iThings/CPSCom 2011
, 10.1109/iThings/CPSCom.2011.12, 558-563, 2011, [URL], This paper targets the students, who are just beginning to engage in research. With the data-mining technologies, using the data of KAKEN (Grant-in-Aid for Scientific Research of Japan), according to students' learning styles and learners' knowledge levels, we propose to create a "Learning by Searching" search engine to provide suitable knowledge and help students to master research trends. "Learning by Searching" provides newly developed pedagogy to meet the knowledge needs of learners..
290. Sachio Hirokawa, Jun Zeng, Comparison of tourism data using double ranking, 1st ACIS International Symposium on Software and Network Engineering, SSNE 2011
Proceedings - 1st ACIS International Symposium on Software and Network Engineering, SSNE 2011
, 10.1109/SSNE.2011.28, 67-72, 2011, [URL], The prompt discovery and mining of user's reputation, opinion and complaint are becoming a hot topic in the field of search engine for Blog and Twitter. This paper proposes "Double Rank" method to analyze the search result using two viewpoints, where the polarity degree of keywords in the search results are evaluated from the viewpoints. Most previous researches concern mainly in positive and negative evaluation as for the polarity degree. Two viewpoints are specified as the search condition in the present paper. The blog articles related to sightseeing are analyzed as case studies, where the characteristics and the sightseeing situation are compared for two prefectures..
291. Jun Zeng, Sachio Hirokawa, Component search engine using tree-view interface for tourist blogs, 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
Proceedings of 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
, 10.1109/ICAwST.2011.6163165, 331-335, 2011, [URL], The existing search engines return the whole web pages as the search results, which make user spend extra time to read the useless information before finding the information they really want. We propose a novel search engine model called "Component Search Engine", which can return the contents satisfying user's query rather than the whole pages. For achieving the purpose, we adopt a Tree-View interface to display the results. Through usability study, we determinate that Component Search Engine using Tree-View interface can improve user's searching experience and efficiency..
292. Xiaobin Wu, Sachio Hirokawa, Chengjiu Yin, Tetsuya Nakatoh, Yoshiyuki Tabata, Extraction and comparison of tourism information on the web, 16th International Symposium on Artificial Life and Robotics, AROB '11
Proceedings of the 16th International Symposium on Artificial Life and Robotics, AROB 16th'11
, 170-173, 2011, The number of tourists to Japan from foreign countries is drastically increased in recent years. However, there is a scene where the traveler is made uneasy by differences between the word, the custom and the culture in traveling abroad. We are aiming at the construction the horizontal search engine intended for tourism information in the Kyushu region as a test case of a special vertical search engine. As the first step for the research, we extracted 312 events from a public tourism portal site and compared the ranking of each event in the site with that in the general search engine. We confirmed a weak correlation of the ranking. Moreover, we found that the big difference of rankings for the events with strong regionality.The number of tourists to Japan from foreign countries is drastically increased in recent years. However, there is a scene where the traveler is made uneasy by differences between the word, the custom and the culture in traveling abroad. We are aiming at the construction the horizontal search engine intended for tourism information in the Kyushu region as a test case of a special vertical search engine. As the first step for these the research, we extracted 312 events from a public tourism portal site and compared the ranking of each event in the site with that in the general search engine. We analyzed the correlation of the rankings and confirmed a weak correlation. In addition, it was also confirmed that there was a large gap for the events with strong regionality..
293. Tetsuya Nakatoh, Chengjiu Yin, Sachio Hirokawa, Extraction and disambiguation of name of place from tourism blogs, 1st ACIS International Symposium on Software and Network Engineering, SSNE 2011
Proceedings - 1st ACIS International Symposium on Software and Network Engineering, SSNE 2011
, 10.1109/SSNE.2011.29, 73-78, 2011, [URL], By development of the Internet in recent years, tourism portal sites and blog articles about tourism increased on WWW. Acquisition of various tourism information became easy. When gathering and classifying the information automatically from blog articles, it is not easy to decide automatically place names used as the key. In this paper, we propose a method of extracting place names from blog articles automatically. Moreover, we also tried disambiguation of a place name..
294. Sachio Hirokawa, Takahiro Baba, Tetsuya Nakatoh, Text mining of bankruptcy information using formal concept analysis, 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
Proceedings of 2011 3rd International Conference on Awareness Science and Technology, iCAST 2011
, 10.1109/ICAwST.2011.6163185, 527-532, 2011, [URL], A lot of information concerning the status of companies are available on the Web. However, a simple search of documents does not explain the meaning or the cause the status. Semantical interpretation and hypotheses generation are necessary for further analysis. This paper proposes a method to analyse the cause and the situation of bankruptcy with respect to particular condition that a user can specify as a query. The method is based on the theory of formal concept analysis. The novelty of the method is in (a) that sentences are considered as objects and words are considered as attributes and (b) that a concise subgraph of the concept lattice is introduced and used to guess the cause. Two cases of interactive and iterative process are shown where a user proceeds from a simple query to a new hypothesis, which would not be able to found by a naive cross tabulation or keyword extraction..
295. Masao Mori, Toshie Tanaka, Sachio Hirokawa, A Document Authoring System for Credible Enterprise Reporting with Data Analysis from Data Warehouse, Proc. SEMAPRO (The Fourth International Conference on Advances in Semantic Processing), pp.218--221, 2010, 2010.12.
296. Kun Qian, Sachio Hirokawa, Kenji Ejima, Xiaoping Du, A fast associative mining system based on search engine and concept graph for large-scale financial report texts, Proc. 2nd IEEE ICIFE ( Information and Financial Engineering), pp.675--679, 2010, 2010.12.
297. Chengjiu Yin, Tetsuya Nakatoh, Sachio Hirokawa, Xiaobin Wu, Jun Zeng, A proposal of search engine XYZ for tourism events, Proc. JCAI (International Joint Conference on Artificial Intelligence), Vol.1, pp.178--181, 2010, 2010.12.
298. Kun Qian, Sachio Hirokawa, Kenji Ejima, Xiaoping Du, A fast associative mining system based on search engine and concept graph for large-scale financial report texts, 2010 2nd IEEE International Conference on Information and Financial Engineering, ICIFE 2010
Proceedings - 2010 2nd IEEE International Conference on Information and Financial Engineering, ICIFE 2010
, 10.1109/ICIFE.2010.5609447, 675-679, 2010.12, [URL], Association mining is widely used in pattern discovery. For large scale financial textual data analysis, however, association mining is relatively less applied due to low efficiency in text manipulation. This paper presents a fast finance textual mining system, based on search engine and concept graph, for large scale financial textual association mining and visualization. Through the experiments on ten years' financial reports of 6,049 companies from NASDAQ and NYSE from 1999 to 2008, it testified that this system could rapidly extracting the characteristic words among millions of texts and visualizing them by concept graph in seconds..
299. Takahiro Baba, Lucing Liu, Sachio Hirokawa, Formal concept analysis of medical incident reports, 14th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2010
Knowledge-Based and Intelligent Information and Engineering Systems - 14th International Conference, KES 2010, Proceedings
, 10.1007/978-3-642-15393-8_24, 6278 LNAI, 207-214, 2010.11, [URL], It is known that a lot of incidents has happened ahead of a serious accident. Such experiences have been collected in medical sites as incident reports. The text mining is expected as a method that discovers the factors of incidents and the improvement of the situation. This paper proposes a method to analyse the co-occurrence relation of the words that appear in the medical incident reports using concept lattice..
300. Park Jong-Hyun, Sachio Hirokawa, Recommender System for Device Sharing in Ubiquitous Environments, 第9回情報科学技術フォーラム(FIT), 2010.09.
301. Yasuhiro Yamada, Sachio Hirokawa, Coloring for Pattern Detection, Proc. 3rd Mahasarakham International Workshop on AI, 27-36, 27-36, 2009.12.
302. Yurie Iino, Sachio Hirokawa, Time series analysis of R&D team using patent information, 13th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, KES 2009
Knowledge-Based and Intelligent Information and Engineering Systems - 13th International Conference, KES 2009, Proceedings
, 10.1007/978-3-642-04592-9_58, 5712 LNAI, 464-471, 2009.12, [URL], Reliable real data is indispensable for the examination, evaluation and the improvement of the organizational structure. This paper proposes a method to use patent documents for analyzing organizational structure of researchers. The method is more efficient and objective compared to personal interview. The structure of research groups is modeled as a "inventors graph", which is a directed graph where each node represents an inventor and an edge represents co-inventor relationship. Empirical evaluation is conducted to cosmetic related companies and their patents that applied between 1998 and 2002 in Japan. It is shown that there is different characteristics in the inventors graph between Japanese companies and foreign companies. Moreover, time series analysis revealed that the inventors graphs of a Japanese company Kao changed in 2001 to foreign company type..
303. Hitoshi Inoue, Koichi Yasutake, Takahiro Sumiya, Osamu Yamakawa, Takahiro Tagawa, Sachio Hirokawa, Visual analysis of online test logs for instructional improvement, 12th IASTED International Conference on Computers and Advanced Technology in Education, CATE 2009
Proceedings of the 12th IASTED International Conference on Computers and Advanced Technology in Education, CATE 2009
, 81-86, 2009.12, Formative assessment is an educational improvement system whose targets are learners, instruction and their evaluation. Quizzes are practical methods to collect basic data to be analyzed in assessment. Online tests of CMS (Course Management System) are expected to help students in taking quizzes as many as times and improve their comprehension. Study logs on CMS are valuable data to understand students' behaviour. The present paper proposes and studies several visualization methods of these logs, not to analyze the students but to analyze the lessons and instruction from view of instructional design..
304. Yurie Iino, Sachio Hirokawa, Time Series Analysis of R&D Team Using Patent Information, Proc. 13th International Conference KES 2009, Part II, Springer LNCS 5712, 464-471, 2009.09.
305. Yurie Iino, Sachio Hirokawa, Evaluation of Research and Development Team Sructure from Patent Documents, 7th International Conference on Patents and Innovation, CD-ROM, 2009.01.
306. 飯野由里江,廣川佐千男, 特許情報に基づく化粧品分野の研究開発体制の分析, 第4回国際シンポジウム日本の技術革新, 43-48, 2008.12.
307. Yurie Iino, Yasuhiro Yamada, Sachio Hirokawa, Structural Analysis of R\&D Division from Patent Documents, IEEE International Conference on e-Business and Engineering, 423-428, 2008.10.
308. Yoshihiro Shimoji, Sachi Hirokawa, Dynamic Thesaurus Construction from English-Japanese Dictionary
, Proc. International Workshop on Ontology Alignment and Visualization, 2008.03.
309. Jong-Hyun Park, Sachio Hirokawa, Ji-Hoon Kang, Extraction of similar information for XML Data, Proc. 7th International Conference on Applications and Principles of Information Science, 2008.01.
310. Yoshihiro Shimoji, Taiki Wada, Sachio Hirokawa, Dynamic thesaurus construction from english-japanese dictionary, CISIS 2008: 2nd International Conference on Complex, Intelligent and Software Intensive Systems
Proceedings - CISIS 2008: 2nd International Conference on Complex, Intelligent and Software Intensive Systems
, 10.1109/CISIS.2008.63, 918-923, 2008, [URL], We propose a method that constructs a hierarchy of words from a set of given documents automatically and dynamically. Given a query, the system retrieves the set of documents that satisfy the query, and the related words are extracted automatically, according their document frequencies. Then a hypernym relations of these related words are obtained using co-occurrence frequencies. Empirical evaluation are conducted for the word hierarchy derived from an English-Japanese dictionary, where the descriptions of a word is considered as one documents. It is shown that meaningful structures are obtained. It is also shown that the proposed method generates more fine hierarchy compared to that obtained by Niwa's and Srinivasan's method..
311. Masao Mori, Tetsuya Nakatoh, Sachio Hirokawa, Links and cycles of web databases, 4th Italian Workshop on Semantic Web Applications and Perspectives, SWAP 2007
SWAP 2007 - 4th Italian Workshop on Semantic Web Applications and Perspectives, Workshop Proceedings
, 314, 2008, This paper proposes a novel framework for composing web databases. Web databases are assumed to have explicit descriptions of I/O attributes and are considered as components of functional compositions. A user writes a script to connect output channels and input channels of components. A script determines a directed graph that may contain cycles which formalizes interactive and iterative behavior of a user through a browser. The interaction and iteration are realised by the notion of CGI-link. Auxiliary filters are introduced as components for universal manipulating tools..
312. Yurie Iino, Yasuhiro Yamada, Sachio Hirokawa, Structural analysis of R and D division from patent documents, IEEE International Conference on e-Business Engineering, ICEBE'08
IEEE International Conference on e-Business Engineering, ICEBE'08 - Workshops: AiR'08, EM2I'08, SOAIC'08, SOKM'08, BIMA'08, DKEEE'08
, 10.1109/ICEBE.2008.29, 423-428, 2008, [URL], This paper describes a method to draw the network of inventors and their technology. It is based on the co-occurrence analysis of inventors of a company described in their patent documents. Empirical evaluation, using 16,375 cosmetic related patent documents, shows that the method discloses the structure of research and development activities of companies..
313. Masao Mori, Tetsuya, Nakatoh, Sachio, Hirokawa, Links and Cycles in Web Databases, Proc. the 4th Workshop on Semantic Web Applications and Perspectives (SWAP 2007), 2007.12.
314. Takahiro Seki, Taiki Wada, Yasuhiro Yamada, Nozomi Ytow, Sachio Hirokawa, Multiple viewed search engine for an e-journal - A case study on zoological science, 12th International Conference on Human-Computer Interaction, HCI International 2007
Human-Computer Interaction
HCI Intelligent Multimodal Interaction Environments - 12th International Conference, HCI International 2007, Proceedings
, 989-998, 2007.12, The multiple viewed search engine presented here retrieves documents of an indicated search area and displays a matrix of the distribution of the clustering from two aspects of the retrieval result. The search engine provides a visual and semantic bird's-eye view of the entire retrieval result. In addition, the characteristic words of each cluster are displayed in the matrix, and supports narrowing of the search. Furthermore, it is possible to immediately change the analysis criteria or the number of clusters and to use a zooming function. Thus, various retrieval conditions for a query can be attempted immediately and continuously. As a case study, this paper performs several analyses on the electronic journal Zoological Science using a multiple viewed search engine..
315. Takahiro Seki, Taiki Wada, Yasuhiro Yamada, Nozomi Ytow, Sachio Hirokawa, Multiple viewed search engine for an e-journal --- a case study on Zoological Science, Proc. The 12th International Conference on Human-Computer Interaction Part IV,
LNCS 4553, pp.989-998, 2007
, 2007.07, The multiple viewed search engine presented here retrieves documents of an indicated search area and displays a matrix of the distribution of the clustering from two aspects of the retrieval result. The search engine provides a visual and semantic bird's-eye view of the entire retrieval result. In addition, the characteristic words of each cluster are displayed in the matrix, and supports narrowing of the search. Furthermore, it is possible to immediately change the analysis criteria or the number of clusters and to use a zooming function. Thus, various retrieval conditions for a query can be attempted immediately and continuously. As a case study, this paper performs several analyses on the electronic journal Zoological Science using a multiple viewed search engine..
316. Masao Mori, Tetsuya Nakatoh, Sachio Hirokawa, Functional Composition of Web Detabases, The 9th International Conference on Asian Digital Libraries, LNCS 4312, pp. 439--448, 2006.11.
317. 中藤哲也, 大森敬介, 廣川佐千男, WebDBのQueryFormにおけるメタデータ自動抽出, 日本データベース学会論文誌 DBSJ Letters, Vol.5, No.2, pp.97--100, 2006.05.
318. Masao Mori, Tetsuya Nakatoh, Sachio Hirokawa, Functional composition of Web databases, 9th International Conference on Asian Digital Libraries, ICADL 2006
Digital Libraries: Achievements, Challenges and Opportunities - 9th International Conference on Asian Digital Libraries, ICADL 2006, Proceedings
, 4312 LNCS, 439-448, 2006, This paper proposes the architecture of the functional composition of Web databases (WebDBs). Unlike a general search engine which receives keywords and returns a list of URLs, a WebDB receives a complex query and returns a list of records. The complex query specifies the condition of each field of the records. The process of composing WebDBs is described as a script, where a user chooses the target WebDBs and describes how to connect the output from one WebDB to the input of another WebDB and how to generate outputs. The novelty of the proposal is that both the WebDBs and output formats are considered as components of the same level and that the reuse of new keywords is represented as a connection (CGI links). Once the process is described as a script, the user can use the script for a new WebDB of his own..
319. Yufen Dou, Eisuke Itoh, Sachio Hirokawa, Daisuke Ikeda, An Approach to Analyzing Correlation between Songs/Artists Using iTMS Playlists, Proc. International Conference on Intelligent Agents, Web Technology and
Internet Commerce (IAWTIC'2005)
, 951-956, 28-30 November 2005, Vienna, 2005.12.
320. Tetsuya Nakatoh, Kensuke Baba, Daisuke Ikeda, Yasuhiro Yamada, Sachio Hirokawa, An Efficient Mapping for Computing the Score of String Matching, Journal of Automata, Languages and Combinatorics, Vol.10, No.5/6, pp.697--704, 2005.11.
321. Tatsuji Kuboyama, Tetsuhiro Miyahara, Sachio Hirokawa, Eisuke Itoh, Information Extraction from Web Pages Using Semi-structured Data Alighment, Proc. 9th World Multi-Conference on Systemics, Cybernetics and Informatics, Vol. I, pp.42--47, 2005.07.
322. Toshiro Minami, Sachio Hirokawa, Towards Multilingual Syllabus Integration, International Journal of Information, Vol.8, No.2, pp.281--290, 2005.03.
323. 篠原正典,廣川佐千男, Web上の高等教育用コンテンツの自動収集と抽出 −シラバスの自動抽出−, 教育システム情報学会誌, Vol.23,No.3, 2005(印刷中), 2005.01.
324. 池田大輔, 山田泰寛, 廣川佐千男, 部分文字列増幅法による共通パタン発見アルゴリズム, 情報処理学会論文誌「数理モデル化と応用」, Vol. 46,No. SIG 2 (TOM 11), pp. 56--66, 2005., 2005.01.
325. Yufeng Dou, Eisuke Ito, Sachio Hirokawa, Daisuke Ikeda, An approach to analyzing correlation between songs/artists using iTMS playlists, International Conference on Computational Intelligence for Modelling, Control and Automation, CIMCA 2005 and International Conference on Intelligent Agents, Web Technologies and Internet Commerce, IAWTIC 2005
Proceedings - International Conference on Computational Intelligence for Modelling, Control and Automation, CIMCA 2005 and International Conference on Intelligent Agents, Web Technologies and Internet
, 1, 951-956, 2005, Digital audio devices have been changing music entertainment environment. Those devices are bundled with music jukebox software, such as Apple's iTunes, Sony's CONNECT player. Jukebox software not only enables us to recode, play, search, purchase music on PC, but also to manage playlist. Anybody can make his/her own playlists, and play music according to the list. In this paper, we focus on iTMS (iTunse Music Stroe) playlists and use them as the data minig resources for a music recommendation system, then developing correlation measuring methods. We have retrieved about 13,000 playlists, and analyzed the frequency of artists/songs, co-occurrence of artists/songs in the playlists. Through the result data, we found out that all graphs we drew are follow the Zipf's law. Furthermore, we have analyzed the hierarchical relation between songs according to their popularity. We proposed a basic idea of popularity measuring method..
326. Tetsuji Kuboyama, Tetsuhiro Miyahara, Sachio Hirokawa, Eisuke Ito, Information extraction from web pages using semi-structured data alignment, 9th World Multi-Conference on Systemics, Cybernetics and Informatics, WMSCI 2005
WMSCI 2005 - The 9th World Multi-Conference on Systemics, Cybernetics and Informatics, Proceedings
, 1, 42-47, 2005, Information extraction from semistructured data such as HTML documents gains importance with the unflagging growth of Web data storage. This paper proposes a structure-based method for extracting Web contents and their metadata from a set of HTML documents generated from a common template, as shown in syllabus and staff data in universities. These HTML documents include a number of grammatical mistakes in HTML, redundant or missing fragments introduced by manual editing. This method first finds a canonical HTML document compliant with the common template. Next, the correspondences of the data between the canonical document and the other documents are identified by an approximate matching algorithm, and aligned according to the correspondences of the data. Experiments have been conducted to extract attribute names for metadata construction and, to align data records from syllabus Web pages in universities..
327. Yasuhiro Yamada, Nick Craswell, Tetsuya Nakatoh, Sachio Hirokawa, Testbed for Information Extraction from Deep Web, Proc. of the 13th
International World Wide Web Conference, Alternate Track Papers and
Posters
, pp.346-347, 2004.05.
328. Yasuhiro Yamada, Nick Craswell, Tetsuya Nakatoh, Sachio Hirokawa, Testbed for information extraction from deep web, 13th International World Wide Web Conference on Alternate Track, Papers and Posters, WWW Alt. 2004
Proceedings of the 13th International World Wide Web Conference on Alternate Track, Papers and Posters, WWW Alt. 2004
, 10.1145/1013367.1013468, 346-347, 2004.05, [URL], Search results generated by searchable databases are served dynamically and far larger than the static documents on the Web. These results pages have been referred to as the Deep Web [1]. We need to extract the target data in results pages to integrate them on different searchable databases. We propose a testbed for information extraction from search results. We chose 100 databases randomly from 114,540 pages with search forms. Therefore, these databases have a good variety. We selected 51 databases which include URLs in a results page and manually identify target information to be extracted. We also suggest evaluation measures for comparing extraction methods and methods for extending the target data..
329. Tetsuya Nakatoh, Yasuhiro Yamada, Sachio Hirokawa,, Automatic Generation of Deep Web Wrappers based on Discovery of Repetition,, Proceeding of the First Asia Information Retrieval Symposium (AIRS
2004)
, pp.269-272, Beijing, China, 2004.01.
330. 山田泰寛,池田大輔,廣川佐千男, 交代数を用いた多言語Webテキストからの共通部分特定とラッパーの生成法, 情報処理学会論文誌, Vol.45, No.9, pp.2138-2145, 2004.01.
331. Yasuhiro Yamada, Tetsuya Nakatoh, Nick Craswell, Sachio Hirokawa, Testbed for information extraction from Deep Web, Thirteenth International World Wide Web Conference Proceedings, WWW2004
Thirteenth International World Wide Web Conference Proceedings, WWW2004
, 1078-1079, 2004, Search results generated by searchable databases are served dynamically and far larger than the static documents on the Web. These results pages have been referred to as the Deep Web [1], We need to extract the target data in results pages to integrate them on different searchable databases. We propose a testbed for information extraction from search results. We chose 100 databases randomly from 114,540 pages with search forms. Therefore, these databases have a good variety. We selected 51 databases which include URLs in a results page and manually identify target information to be extracted. We also suggest evaluation measures for comparing extraction methods and methods for extending the target data..
332. S. Hirokawa, E. Itoh, T. Miyahara, Semi-Automatic Construction of Metadata from A Series of Web Documents, 16th Australian Joint Conference on Artificial Intelligence, 2903, 942-953, 2003.12.
333. M. Noguchi, S. Hirokawa, A Prototype of Search Engine for Tables on the Web, pp.561-564, 2003.11.
334. Y. Matsunaga, S. Yamada, E. Ito, S. Hirokawa, A Web Syllabus Crawler and its Efficiency Evaluation, Proc. ISEE 2003, pp. 565-568, 2003.11.
335. T. Matsuo, S. Hirokawa, Generation and Expansion of Web Navigation from Browsing Log, Proc. ISEE 2003, 2003.11.
336. T. Nakatoh, K. Ohmori, Y. Yamada, S. Hirokawa, Generation of Metadata for Complex Query Forms, Proc. ISEE 2003, pp.291-294, 2003.11.
337. D. Ikeda, S. Hirokawa, Y. Yamada, Pattern Discovery of Genome Sequences by Substring Amplification, Proc. ISEE 2003, pp.637-640, 2003.11.
338. T. Nakatoh, K. Baba, D. Ikeda, Y. Yamada, S. Hirokawa, An Efficient Mapping for scores of String Matching, Prague Stringology Conference '03, http://cs.felk.cvut.cz/psc/, 2003.09.
339. D. Ikeda, Y. Yamada, S. Hirokawa, Expressive power of tree and string based wrappers, Proc. IJCAI Workshop on Information Integration on the Web, pp.21-26,2003, 2003.08.
340. T. Miyahara, Y. Suzuki, T. Shoudai, T. Uchida, S. Hirokawa, K. Takahashi, H. Ueda, Extraction of Tag Tree Patterns with Contractible Variables from Irregular Semistructured data, Proc. PAKDD, 2637, 430-436, pp.430--436, 2003.04.
341. 池田大輔,山田泰寛,廣川佐千男, Web上の多言語テキストデータからのラッパー生成, 九州大学情報基盤センター 情報基盤センター年報, Vol 3, pp.7-14, 2003.03.
342. T. Nakatoh, K. Ohmori, Y. Yamada, S. Hirokawa, Complex Query and Metadata, Proc. ISEE 2003, pp.291-294, 2003.01.
343. 山田信太郎, 松永吉広, 伊東栄典, 廣川佐千男, Webシラバス情報収集エージェントの試作, 電子情報通信学会論文誌D1, J86-D1(8),pp.566-574, 2003.01.
344. Tetsuhiro Miyahara, Yusuke Suzuki, Takayoshi Shoudai, Tomoyuki Uchida, Sachio Hirokawa, Kenichi Takahashi, Hiroaki Ueda, Extraction of tag tree patterns with contractible variables from irregular semistructured data, 7th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2003
Advances in Knowledge Discovery and Data Mining
, 2637, 430-436, 2003.01, Information Extraction from semistructured data becomes more and more important. In order to extract meaningful or interesting contents from semistructured data, we need to extract common structured patterns from semistructured data. Many semistructured data have irregularities such as missing or erroneous data. A tag tree pattern is an edge labeled tree with ordered children which has tree structures of tags and structured variables. An edge label is a tag, a keyword or a wildcard, and a variable can be substituted by an arbitrary tree. Especially, a contractible variable matches any subtree including a singleton vertex. So a tag tree pattern is suited for representing common tree structured patterns in irregular semistructured data. We present a new method for extracting characteristic tag tree patterns from irregular semistruc-tured data by using an algorithm for finding a least generalized tag tree pattern explaining given data. We report some experiments of applying this method to extracting characteristic tag tree patterns from irregular semistructured data..
345. Sachio Hirokawa, Eisuke Ito, Tetsuhiro Miyahara, Semi-automatic construction of metadata from a series of web documents, 16th Australian Conference on Artificial Intelligence, AI 2003
AI 2003
Advances in Artificial Intelligence - 16th Australian Conference on AI, Proceedings
, 942-953, 2003.01, Metadata plays an important role in discovering, collecting, extracting and aggregating Web data. This paper proposes a method of constructing metadata for a specific topic. The method uses Web pages that are located in a site and are linked from a listing page. Web pages of recipes, real estates, used cars, hotels and syllabi are typical examples of such pages. We call them a series of Web documents. A series of Web pages have the same appearance when a user views them with a browser, because it is often the case that they are written with the same tag pattern. The method uses the tag-pattern as the common structure of the Web pages. Individual contents of the pages appear as plain texts embedded between two consecutive tags. If we remove the tags, it becomes a sequence of plain texts. The plain texts in the same relative position can be interpreted as attribute values if we presume that the pages represent records of the same kind. Most of these plain texts in the same position vary page to page. But, it may happen that the same texts show up at the same relative position in almost all pages. These constant texts can be considered as attribute names. “Location”, “Rating” and “Travel from Airport” are examples of such constant texts for pages of hotel information. If the frequency of a text is higher than a threshold, we accept it as a component of metadata. If we mark a constant text with “N” and a variable text with “V”, the sequence of plain texts forms a series of N’s and V’s. A page in a series contain two kinds of NV sequence pattern. The first pattern is (NV)n, which we call vertical, where an attribute value follows the attribute name immediately. The second pattern is NnVn, which we call horizontal, where names occur in the first row and the same number of values follow in the next row. Thus we can understand the meaning of values and can construct records from a series of Web pages..
346. Yasuhiro Yamada, Daisuke Ikeda, Sachio Hirokawa, Automatic Wrapper Generation for Multilingual Web Resources, Proceedings of the Fifth International Conference on Discovery Science,
Lecture Notes in Computer Science, Springer-Verlag
, 2534, 332-339, Vol.2534, pp.332-339, 2002.11.
347. 山田信太郎, 松永吉広, 伊東栄典, 廣川佐千男, Webシラバス情報収集エージェントの試作, エージェント合同シンポジウム (JAWS 2002), pp.371-378, 2002.11.
348. 山田信太郎, 伊東栄典, 廣川佐千男, Web上に公開されたシラバス情報の自動収集, 情報処理学会 マルチメディア,分散,協調とモバイル (DICOMO 2002) シンポジウム論文集, pp.137-140, 2002.07.
349. 樺島結城, 廣川佐千男, リンク情報に基づく検索エンジンの比較, 第13回データ工学ワークショップ予稿集, http://www.ieice.org/iss/de/DEWS/proc/2002/proceedings.html, 2002.03.
350. 酒井美由紀, 廣川佐千男, 検索サイトラッパー検証のための検索結果件数推定方法, 第13回データ工学ワークショップ予稿集, http://www.ieice.org/iss/de/DEWS/proc/2002/proceedings.html, 2002.03.
351. 渡辺精一郎, 廣川佐千男, 研究会プログラムのWebページからのデータ抽出, 第13回データ工学ワークショップ予稿集, http://www.ieice.org/iss/de/DEWS/proc/2002/proceedings.html, 2002.03.
352. T. Nakatoh, Y. Koga, A. Uhl, S. Hirokawa,, Automatic Estimation of Query Syntax for Search Sites, Proceedings of PYIWIT'02 (Pan-Yellow-Sea International Workshop on
Information Technologies for Network Era),
, pp.329-332, 2002.01.
353. Sachio Hirokawa, Daisuke Ikeda, Visualization and Analysis of Web Graphs, Springer LNCS, vol.2281, pp.616-627, 2002.01.
354. Yasuhiro Yamada, Daisuke Ikeda, Sachio Hirokawa, Automatic wrapper generation for multilingual web resources, 5th International Conference on Discovery Science, DS 2002
Discovery Science - 5th International Conference, DS 2002, Proceedings
, 332-339, 2002.01, We present a wrapper generation system to extract contents of semi-structured documents which contain instances of a record. The generation is done automatically using general assumptions on the structure of instances. It outputs a set of pairs of left and right delimiters surrounding instances of a field. In addition to input documents, our system also receives a set of symbols with which a delimiter must begin or end. Our system treats semi-structured documents just as strings so that it does not depend on markup and natural languages. It does not require any training examples which show where instances are. We show experimental results on both static and dynamic pages which are gathered from 13 Web sites, markuped in HTML or XML, and written in four natural languages. In addition to usual contents, generated wrappers extract useful information hidden in comments or tags which are ignored by other wrapper generation algorithms. Some generated delimiters contain whitespaces or multibyte characters..
355. Sachio Hirokawa, Daisuke Ikeda, Visualization and analysis of Web Graphs, Progress in Discovery Science, 2281, 616-627, 2002, We review the progress ofour research on Web Graphs. A Web Graph is a directed graph whose nodes are Web pages and whose edges are hyperlinks between pages. Many people use bookmarks and pages oflinks as a knowledge on internet. We developed a visualization system ofW eb Graphs. It is a system for construction and analysis of Web graphs. For constructing and analysis oflarge graphs, the SVD (Singular Value Decomposition) ofthe adjacency matrix ofthe graph is used. The experimental application ofthe system yield some discovery that are unforseen by other approach. The scree plots of the singular values ofthe adjacency matrix is introduced and confirmed that can be used as a measure to evaluate the Web space..
356. S. Hirokawa, S. Watanabe, Y. Koga, T. Taguchi, Automatic Feature Extraction of Search Sites, Proc. SSGRR 2001, CD-ROM, 2001.01.
357. Kengo Nishino, Daisuke Nagano, Sachio Hirokawa, Generation of Navigation Script from Log and Link, Proc. WebNet2001, pp.534-539, 2001.01.
358. T. Nakatoh, M. Sakai, Y. Koga, S. Hirokawa, Generation of Query URL for Search Sites, Proc. SSGRR 2001 Winter, 2001.01.
359. Kengo Nishino, Sachio Hirokawa, Daisuke Nagano, Rapid Prototyping of WWW Tour from Browsing History and Link, Proc. ICCE2001, pp.800-8003, 2001, 2001.01.
360. Sachio Hirokawa, Daisuke Ikeda, Webグラフの構造解析, 人工知能学会誌, vol.16(4), p.525-529, 2001.01.
361. Kensuke Baba, Sachio Hirokawa, Ken Etsu Fujita, Parallel reduction in type free λμ-calculus, Computing: The Australasian Theory Symposium (CATS 2001)
Electronic Notes in Theoretical Computer Science
, 10.1016/S1571-0661(04)80878-8, 42, 52-66, 2001.01, [URL], The typed λμ-calculus is known to be strongly normalizing and weakly Church-Rosser, and hence becomes confluent. In fact, Parigot formulated a parallel reduction to prove confluence of the typed λμ-calculus by "Tait-and-Martin-Löf" method. However, the diamond property does not hold for his parallel reduction. The confluence for type-free λμ-calculus cannot be derived from that of the typed λμ-calculus and is not confirmed yet as far as we know. We analyze granularity of the reduction rules, and then introduce a new parallel reduction such that both renaming reduction and consecutive structural reductions are considered as one step parallel reduction. It is shown that the new formulation of parallel reduction has the diamond property, which yields a correct proof of the confluence for type free λμ-calculus. The diamond property of the new parallel reduction is also applicable to a call-by-value version of the λμ-calculus containing the symmetric structural reduction rule..
362. Yasuhiro Yamada, Daisuke Ikeda, Sachio Hirokawa, SCOOP
A record extractor without knowledge on input, 4th International Conference on Discovery Science, DS 2001
Discovery Science - 4th International Conference, DS 2001, Proceedings
, 482-487, 2001.01, We present a record extractor system SCOOP. We assume that semi-structured documents given to SCOOP contain similar formats and each of them has only a record consisting of some different fields. SCOOP treats a document as just a string and does not use knowledge on input except that a field is surrounded with delimiters, a left delimiter ends with “>”, and the corresponding right delimiter begins with “<”. By counting substrings, SCOOP roughly divides into two parts: contents of the fields and others. SCOOP counts substrings near boundaries of two parts and extracts the most frequent substrings as delimiters. We show experimental results with news articles written in English or Japanese. A record consists of the headline and the body text on this experiment. SCOOP extracts records at a high rate..
363. Daisuke Ikeda, Yasuhiro Yamada, Sachio Hirokawa, Eliminating useless parts in semi-structured documents using alternation counts, 4th International Conference on Discovery Science, DS 2001
Discovery Science - 4th International Conference, DS 2001, Proceedings
, 2226, 113-127, 2001, We propose a preprocessing method for Web mining which, given semi-structured documents with the same structure and style, distinguishes useless parts and non-useless parts in each document without any knowledge on the documents. It is based on a simple idea that any n-gram is useless if it appears frequently. To decide an appropriate pair of length n and frequency a, we introduce a new statistic measure alternation count. It is the number of alternations between useless parts and non-useless parts. Given news articles written in English or Japanese with some non-articles, the algorithm eliminates frequent n-grams used for the structure and style of articles and extracts the news contents and headlines with more than 97% accuracy if articles are collected from the same site. Even if input articles are collected from different sites, the algorithm extracts contents of articles from these sites with at least 95% accuracy. Thus, the algorithm does not depend on the language, is robust for noises, and is applicable to multiple formats..
364. Daisuke Ikeda, Sachio Hirokawa, Extracting Positve and Negative Keywords for Web Communities, Springer LNCS, vol.1967, pp.299-303, 2000.01.
365. Sachio Hirokawa, Daisuke Ikeda, Tuyoshi Taguchi, Generation of Knowledge Network from Link Information, Discovery Science and Data Mining, in: S. Morishita, S. Miyano Eds,
pp.272-281, 2000.01.
366. T. Taguchi, Y. Koga, S. Hirokawa, Integration of Search Sites of the World Wide Web, Proceedings of International Forum cum Conference on Information Technology and Communication, vol.2, pp.25-32, 2000.01.
367. Sachio Hirokawa, Kengo Nishino, Daisuke Nagano,, Navigation Script for the World Wide Web, Proc. ICCE/ICCAI2000, pp.800-803, 2000.01.
368. Sachio Hirokawa, Yuichi Komori, Misao Nagayama, A lambda proof of the P-W theorem, Journal of Symbolic Logic, 10.2307/2695080, 65, 4, 1841-1849, 2000.01, [URL], The logical system P-W is an implicational non-commutative intuitionistic logic defined by axiom schemes B = (b → c) → (a → b) → a → c. B′ = (a → b) → (b → c) → a → c. I = a → a with the rules of modus ponens and substitution. The P-W problem is a problem asking whether α = β holds if α → β and β → α are both provable in P-W. The answer is affirmative. The first to prove this was E. P. Martin by a semantical method. In this paper, we give the first proof of Martin's theorem based on the theory of simply typed λ-calculus. This proof is obtained as a corollary to the main theorem of this paper, shown without using Martin's Theorem, that any closed hereditary right-maximal linear (HRML) λ-term of type α → α is βη-reducible to λx.x. Here the HRML λ-terms correspond, via the Curry-Howard isomorphism, to the P-W proofs in natural deduction style..
369. Daisuke Ikeda, Sachio Hirokawa, Extracting positive and negative keywords for web communities, 3rd International Conference on Discovery Science, DS 2000
Discovery Science - 3rd International Conference, DS 2000, Proceedings
, 1967, 299-303, 2000.
370. Daisuke Ikeda, Tuyoshi Taguchi, Sachio Hirokawa, Developing a Knowledge Network of URLs, Springer LNCS, 1721, 328-329, vol.1721, pp.328-329, 1999.01.
371. Daisuke Ikeda, Tsuyoshi Taguchi, Sachio Hirokawa, Developing a knowledge network of URLs, 2nd International Conference on Discovery Science, DS 1999
Discovery Science - 2nd International Conference, DS 1999, Proceedings
, 10.1007/3-540-46846-3_34, 328-329, 1999.01, [URL].
372. Sachio Hirokaw, Infiniteness of Proof(alpha) is Polynomial-Space Complete, Theor. Comput. Sci., 10.1016/S0304-3975(97)00168-0, 206, 1-2, 331-339, 1998.07.
373. Toshiro Minami, Hideto Sazuka, Sachio Hirokawa, Takeshi Ohtani, Living with ZK - An Approach towards Communication with Analogue Messages, Proc. KES98, pp.369-374, 1998.01.
374. Sachio Hirokawa, Tsuyoshi Taguchi, KN on ZK — Knowledge network on network note pad ZK, 1st International Conference on Discovery Science, DS 1998
Discovery Science - 1st International Conference, DS 1998, Proceedings
, 1532, 411-412, 1998.
375. Toshiro Minami, Hideto Sazuka, Sachio Hirokawa, Takeshi Ohtani, Living with ZK - an approach towards communication with analogue messages, Unknown Journal, 1, 369-374, 1998, It is well known that we communicate more with non-verbal means than verbal ones. It is also true when we communicate with drawing using paper and pencils. We use not only words and texts but also use marks, symbols, and other non-verbal information like arrangement or colors. This paper proposes a new approach to electronic communication with verbal messages together with non-verbal, or analogue, messages. ZK(ZeichenblocK) provides us a means of exchanging analogue messages in a network-transparent way. Its potential importance is demonstrated with applications to such as `analogue questionnaire' and `concept representation.' Such analogue data which is represented as ZK-scripts are effectively distributed in the `word-of-mouth' mechanism..
376. Masako Takahashi, Yohji Akama, Sachio Hirokaw, Normal Proofs and Their Grammar, Inf. Comput, 10.1006/inco.1996.0027, 125, 2, 144-153, 1996.07.
377. Sachio Hirokawa, The proofs of α → α in P - W, Journal of Symbolic Logic, 61, 1, 195-211, 1996.03, The syntactic structure of the system of pure implicational relevant logic P - W is investigated. This system is defined by the axioms B = (b → c) → (a → b) → a → c, B′ = (a - b) → (b → c) → a → c, I = a → a, and the rules of substitution and modus ponens. A class of λ-terms, the closed hereditary right-maximal linear λ-terms, and a translation of such λ-terms M to BB′ I-combinators M+ is introduced. It is shown that a formula α is provable in P - W if and only if α is a type of some λ-term in this class. Hence these λ-terms represent proof figures in the Natural Deduction version of P - W. Errol Martin (1982) proved that no formula with form α → α is provable in P - W without using the axiom I. We show that a β-normal form λ-term M in the class is η reducible to λx.x if the translated BB′ I-combinator M+ contains I. Using this theorem and Martin's result, we prove that a λ-term in the class is βη-reducible to λx.x if the λ-term has a type α → α. Hence the structure of proofs of α → α in P - W is determined..
378. Sachio Hirokawa, Yuichi Komori, Izumi Takeuti, A Reduction Rule for Peirce Formula, Studia Logica, 10.1007/BF00372774, 56, 3, 419-426, 1996.01, [URL], A reduction rule is introduced as a transformation of proof figures in implicational classical logic. Proof figures are represented as typed terms in a λ-calculus with a new constant p((α→β)→α)→α. It is shown that all terms with the same type are equivalent with respect to β-reduction augmented by this P-reduction rule. Hence all the proofs of the same implicational formula are equivalent. It is also shown that strong normalization fails for βP-reduction. Weak normalization is shown for βP-reduction with another reduction rule which simplifies α of ((α → β) → α) → α into an atomic type..
379. M. Takahashi, Y. Akama, Sachio Hirokawa, Normal proofs and their grammar, 2nd International Symposium on Theoretical Aspects of Computer Software, TACS 1994
Theoretical Aspects of Computer Software - International Symposium TACS 1994, Proceedings
, 465-493, 1994.01, First we give a grammatical (or equational) description of the set {M normal form │ Γ ⊢ M : A} for a given basis Γ and a given type A in the simple type system, and give some applications of the description. Then we extend the idea to systems in λ-cube and more generally to normalizing pure type systems. The attempt resulted in derived (or ‘macro’) rules the totality of which is sound and complete for type assignments of normal terms. A feature of the derived rules is that they reflect the syntactic structure of legal terms in normal form, and thus they may give us more global view than the original definition of the systems..
380. Sachio Hirokawa, Principal Types of BCK-lambda-Terms, Theor. Comput. Sci., 10.1016/0304-3975(93)90171-O, 107, 2, 253-276, 1993.07.
381. Yuichi Komori, Sachio Hirokawa, The Number of Proofs for a BCK-Formula, J. Symb. Log., 10.2307/2275222, 58, 2, 626-628, 1993.07.
382. Sachio Hirokawa, The relevance graph of a BCK-formula, Journal of Logic and Computation, 10.1093/logcom/3.3.269, 3, 3, 269-285, 1993.06, [URL], It is known that the set of BCK-formulas which is provable by the detachment rule of Meredith is identical to the set pts(BCK) of principal type-schemes of BCK-λ-terms. This paper shows a characterization of the set pts(BCK-β) of principal type-schemes of BCK-λ-terms in β-normal form. To characterize the set pts(BCK), a 'relevance relation' is defined between type variables in a type. A type variable b is relevant to a type variable c in a type α iff α contains a negative occurrence of a subtype of the form (... → b) → ... → c. The relevance graph G(α) of the type α is the directed graph induced by this relevance relation. A type variable is said to be positive iff it occurs in a positive position and negative otherwise. It is proved that a type α is in pts(BCK-β) iff α satisfies: (a) every type variable occurs exactly once in a negative position and at most once in a positive position; (b) no negative type variable is relevant to any type variable but itself and the subgraph of G(α) whose nodes are positive type variables of α is a tree whose root is the rightmost type variable in α; (c) each positive type variable in a subtype γ is relevant to the right-most type variable in γ..
383. Sachio Hirokawa, Balanced Formulas, BCK-Minimal Formulas and Their Proofs, Proc. LFCS 1992, 10.1007/BFb0023874, 620, 198-208, 1992.07.
384. Sachio Hirokawa, The converse principal type-scheme theorem in lambda calculus, Studia Logica, 10.1007/BF00370332, 51, 1, 83-95, 1992.03, [URL], A principal type-scheme of a λ-term is the most general type-scheme for the term. The converse principal type-scheme theorem (J.R. Hindley, The principal typescheme of an object in combinatory logic, Trans. Amer. Math. Soc.146 (1969) 29-60) states that every type-scheme of a combinatory term is a principal type-scheme of some combinatory term. This paper shows a simple proof for the theorem in λ-calculus, by constructing an algorithm which transforms a type assignment to a λ-term into a principal type assignment to another λ-term that has the type as its principal type-scheme. The clearness of the algorithm is due to the characterization theorem of principal type-assignment figures. The algorithm is applicable to BCIW-λ-terms as well. Thus a uniform proof is presented for the converse principal type-scheme theorem for general λ-terms and BCIW-λ-terms..
385. Sachio Hirokawa, Prinipal Type Assignment to Lambda Terms, Int. J. Found. Comput. Sci., 2, 2, 149-162, 1991.07.
386. Sachio Hirokawa, BCK-formulas having unique proofs, 4th Biennial Summer Conference on Category Theory and Computer Science, 1991
Category Theory and Computer Science, Proceedings
, 10.1007/BFb0013460, 106-120, 1991.01, [URL], The set of relevantly balanced formulas is introduced in implicational fragment of BCK-logic. It is shown that any relevantly balanced formula has unique normal form proof. Such formulas are defined by the ‘relevance relation’ between type variables in a formula. The set of balanced formulas (or equivalently one-two-formulas) is included in the relevantly balanced formulas. The uniqueness of normal form proofs is known for balanced formulas as the coherence theorem. Thus the result extends the theorem with respect to implicational formulas. The set of relevantly balanced formulas is characterized as the set of irrelevant substitution instances of principal type-schemes of BCK-λ-terms..
387. Sachio Hirokawa, Principal type-schemes of BCI-lambda-terms, 1st International Conference on Theoretical Aspects of Computer Software, TACS 1991
Theoretical Aspects of Computer Software - International Conference TACS 1991, Proceedings
, 10.1007/3-540-54415-1_68, 633-650, 1991.01, [URL], A BCI-λ-term is a λ-term in which each variable occurs exactly once. It represents a proof figure for implicational formula provable in linear logic. A principal type-scheme is a most general type to the term with respect to substitution. The notion of “relevance relation” is introduced for type-variables in a type. Intuitively an occurrence of a type-variable b is relevant to other occurrence of some type-variable c in a type α, when b is essentially concerned with the deduction of c in α. This relation defines a directed graph G(α) for type-variables in the type. We prove that a type a is a principal type-scheme of BCI-λ-term iff (a), (b) and (c) holds: (a) Each variable occurring in α occurs exactly twice and the occurrences have opposite sign. (b) G(α) is a tree and the right-most type variable in α is its root. (c) For any subtype γ of α, each type variable in γ is relevant to the right-most type variable in γ. A type-schemes of some BCI-λ-term is minimal iff it is not a non-trivial substitution instance of other type-scheme of BCI-λ-term. We prove that the set of BCI-minimal types coincides with the set of principal type-schemes of BCI-λ-terms in βη-normal form..
388. Shoji Sekimoto, Sachio Hirokawa, One-step recurrent terms in λ-β-calculus, Theoretical Computer Science, 10.1016/0304-3975(88)90079-5, 56, 2, 223-231, 1988.01, [URL], A necessary and sufficient condition for cycling reductions to be recurrent is given. A one-step recurrent term is a term in λ-β-calculus whose one-step reductums are all reducible to the term. It is a weakened notion of minimal form or recurrent term in the λ-β-calculus. In this note, a one-step recurrent term which is not recurrent is shown. That term becomes a counter- example for a conjecture presented by Klop. By analysis of the reduction cycles of one-step recurrent terms, a necessary and sufficient condition for a one-step recurrent term to be recurrent is given..
389. Sachio Hirokawa, Complexity of the combinator reduction machine, Theoretical Computer Science, 10.1016/0304-3975(85)90076-3, 41, C, 289-303, 1985.01, [URL], The complexity of the computation of recursive programs by the combinator reduction machine is studied. The number of the reduction steps in compared between the two models of computation. The main theorem states that the time required by the reduction machine is linear in that of the program scheme. The coefficient of the linearity was shown to be O(n2), where n is the maximal number of variables of the functions being used. For the analysis of the combinator codes, the notion of extended combinator code is introduced..

九大関連コンテンツ

pure2017年10月2日から、「九州大学研究者情報」を補完するデータベースとして、Elsevier社の「Pure」による研究業績の公開を開始しました。
 
 
九州大学知的財産本部「九州大学Seeds集」