記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Learning analytics, blending education theory, psychology, statistics, and computer science, utilizes data about learners and their environments to enhance education. Artificial Intelligence advances this field by personalizing learning and providing predictive insights. However, the opaque ’black box’ nature of AI decision-making poses challenges to trust and understanding within educational settings. This paper presents a novel visual analytics method to predict whether a student is at risk of failing a course. The proposed method is based on a dendritic neuron model (DNM), which not only performs excellently in prediction, but also provides an intuitive visual presentation of the importance of learning behaviors. It is worth emphasizing that the proposed DNM has a better performance than recurrent neural network (RNN), long short term memory network (LSTM), gated recurrent unit (GRU), bidirectional long short term memory network (BiLSTM) and bidirectional gated recurrent unit (BiGRU). The powerful prediction performance can assist instructors in identifying students at risk of failing and performing early interventions. The importance analysis of learning behaviors can guide students in the development of learning plans.

DOI： 10.1007/978-981-97-5495-3_14

Web of Science

Scopus

researchmap

出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

In actual university education, students’ performance prediction is important for assessing their mastery of specific knowledge areas and providing feedback. To address the limitations of existing grade or at-risk predictions that fail to take students’ knowledge mastery into account, we propose a model, question&answer knowledge attention for exam performance prediction (QAKAP) and focus on predicting students’ performance on each final exam question (binary prediction). This model consists of three key modules: a Topic-Aware Attention Module, a Knowledge Mastery Estimation Module, and a Prediction Module. By integrating natural language processing (NLP) methods and the attention mechanism, our model can capture the relevance of questions from question text data and then generate the student embeddings to predict student performance on each final exam question. The primary contribution of this work is the development of a general prediction model for real university education, applicable to students’ exam performance predictions. We collected a dataset from 494 students across four undergraduate courses, including the question text data and students’ performance on these exam questions. The results show that QAKAP outperforms other machine-learning methods, demonstrating its effectiveness in predicting students’ exam performance.

DOI： 10.1007/978-3-031-72315-5_26

Scopus

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Communications in Computer and Information Science  

This study investigates the effects of generative AI on the knowledge comprehension of university students, focusing on the use of ChatGPT strategies. Data from 81 junior students who used the ChatGPT worksheet were collected and analyzed. Path analysis revealed complex interactions between ChatGPT strategy use, e-book reading behaviors, and students’ prior perceived understanding of concepts. Students’ prior perceived understanding and reading behaviors indirectly affected their final scores, mediated by the ChatGPT strategy use. The mediation effects indicated that reading behaviors significantly influenced final scores through ChatGPT strategies, indicating the importance of the interaction with learning materials. Further regression analysis identified the specific ChatGPT strategy related to verifying and comparing information sources as significantly influenced by reading behaviors and directly affecting students’ final scores. The findings provide implications for practical strategic guidance for integrating ChatGPT in education.

DOI： 10.1007/978-3-031-64315-6_12

Scopus

researchmap

出版者・発行元：CEUR Workshop Proceedings  

With the advent of generative artificial intelligence (AI), the scope of data analysis, prediction of performances, real-time feedback, etc. in learning analytics has widened. The purpose of this study is to explore the possibility of using generative AI to analyze educational data. Moreover, the performances of two large language models (LLMs): GPT-4 and text-davinci-003, are compared with respect to different types of analyses. Additionally, a framework, LangChain, is integrated with the LLM in order to achieve deeper insights into the analysis, which can be beneficial for beginner data scientists. LangChain has a component called an agent, which can help study the analysis being performed step-by-step. Furthermore, the impact of the OpenLA library, which pre-processes the data by calculating the number of reading seconds of students, counting the number of operations performed by students, and making page-wise behavior of each student, is also studied. Besides, factors with the most significant impact on students’ performances were also discovered in this analysis. The results show that GPT-4, when using the data pre-processed by OpenLA, provides the best analysis in terms of both, the accuracy of the final answer, and the step-by-step insights provided by LangChain’s agent. Also, we learn the significance of reading time and interactions used (Add marker, bookmark, memo) by students in predicting grades.

Scopus

掲載種別：研究論文（国際会議プロシーディングス）  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/lak/2024w

掲載種別：研究論文（国際会議プロシーディングス）  

In recent years, due to the impact of Coronavirus disease (COVID-19), digital platforms have developed rapidly and accumulated a large amount of data. To better utilize the comprehensive and diverse data stored in online platforms for data mining, such as learning behavior analysis or performance prediction, and to provide guidance and valuable feedback for educator became more important. For the current analysis of learning behaviors by time series data with DNN method, the interpretability is not enough. This paper proposes a method based on the simultaneous use of learning behaviors and learning materials to obtain the representation of learned knowledge, and through multiple cross-validations, the effect of this knowledge representation has a certain improvement on the original data, and the interpretability can promote the feedback function.

Scopus

researchmap

出版者・発行元：CEUR Workshop Proceedings  

In recent years, due to the impact of Coronavirus disease (COVID-19), digital platforms have developed rapidly and accumulated a large amount of data. To better utilize the comprehensive and diverse data stored in online platforms for data mining, such as learning behavior analysis or performance prediction, and to provide guidance and valuable feedback for educator became more important. For the current analysis of learning behaviors by time series data with DNN method, the interpretability is not enough. This paper proposes a method based on the simultaneous use of learning behaviors and learning materials to obtain the representation of learned knowledge, and through multiple cross-validations, the effect of this knowledge representation has a certain improvement on the original data, and the interpretability can promote the feedback function.

Scopus

掲載種別：研究論文（学術雑誌）   出版者・発行元：IEEE Access  

Student performance prediction was deployed to predict learning performance to identify at-risk students and provide interventions for them. However, prediction models should also consider external factors along with learning activities, such as course duration. Thus, we aim to distinguish the difference factor between the time dimension (duration of the course) and the feature dimension (students' learning activities) by attention weights to provide helpful information and improve predictions of student performance. In this study, we introduce Attention-Based Artificial Neural Network (Attn-ANN), a novel model in educational data mining. The Attn-ANN combines attention weighting on the time and feature dimensions to examine the significance of lectures and learning activities and makes predictions by visualizing attention weight. We found that the Attn-ANN had a better area under the curve scores than conventional algorithms, and the attention mechanism allowed models to focus on input selectively. Incorporating the attention weighting of both the time and feature dimensions improved the prediction performance in an ablation study. Finally, we investigated and analyzed the model's decision, finding that the Attn-ANN may be able to create synergy in real-world scenarios between the Attn-ANN's predictions and instructors' expertise, which underscores a novel contribution to engineering applications for interventions for at-risk students.

DOI： 10.1109/ACCESS.2024.3429554

Web of Science

Scopus

researchmap

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

出版者・発行元：31st International Conference on Computers in Education, ICCE 2023 - Proceedings  

Understanding students' coding behaviors is crucial for providing targeted support in programming education. Automatic analysis of coding behaviors using machines can address the limitations of manual monitoring. Previous studies focused on coding behavior representations without considering differences relative to a model answer. We propose embedding vectors that capture these differences, enabling the distinction between simple and complex code solutions. Evaluating these vectors by predicting assignment scores, we achieved over 15% higher accuracy compared to conventional methods. This approach has the potential to enhance teachers' understanding of students' coding behaviors and improve support in programming education.

Scopus

掲載種別：研究論文（国際会議プロシーディングス）  

Understanding students' coding behaviors is crucial for providing targeted support in programming education. Automatic analysis of coding behaviors using machines can address the limitations of manual monitoring. Previous studies focused on coding behavior representations without considering differences relative to a model answer. We propose embedding vectors that capture these differences, enabling the distinction between simple and complex code solutions. Evaluating these vectors by predicting assignment scores, we achieved over 15% higher accuracy compared to conventional methods. This approach has the potential to enhance teachers' understanding of students' coding behaviors and improve support in programming education.

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語  

DOI： 10.1111/1759-7714.15014

Web of Science

PubMed

記述言語：その他   掲載種別：研究論文（学術雑誌）   出版者・発行元：Journal of Educational Computing Research  

This study helps improve the early prediction of student performance by RNN-FitNets, which applies knowledge distillation (KD) to the time series direction of the recurrent neural network (RNN) model. The RNN-FitNets replaces the teacher model in KD with “an RNN model with a long-term time-series in which the features during the entire course are inputted” and the student model in KD with “an RNN model with a short-term time-series in which only the features during the early stages are inputted.” As a result, the RNN model in the early stage was trained to output the same results as the more accurate RNN model in the later stages. The experiment compared RNN-FitNets with a normal RNN model on a dataset of 296 university students in total. The results showed that RNN-FitNets can improve early prediction. Moreover, the SHAP value was employed to explain the contribution of the input features to the prediction results by RNN-FitNets. It was shown that RNN-FitNets can consider the future effects of the input features from the early stages of the course.

DOI： 10.1177/07356331221129765

Web of Science

Scopus

researchmap

記述言語：その他   掲載種別：研究論文（学術雑誌）   出版者・発行元：IEEE Transactions on Learning Technologies  

In this study, we propose an integrated system to support learners' reviews. In the proposed system, the review dashboard is used to recommend review contents that are adaptive to the individual learner's level of understanding and to present other information that is useful for review. The pages of the digital learning materials that are estimated to be insufficiently understood by each learner and the webpages related to those pages are recommended. As a method for estimating such pages, we consider extracting the pages related to the questions that were answered incorrectly. We examined the accuracy of matching each question with the pages of the learning materials. We also conducted an experiment to verify the usefulness of the system and its effect on learning using a review dashboard. In the experiment, the evaluation of the review dashboard indicated that at least half of the participants found it useful for most types of feedback. In addition, the rate of change in quiz scores was significantly higher in the group using the review dashboard, which indicates that using the review dashboard has the effect of improving learning.

DOI： 10.1109/TLT.2022.3225206

Web of Science

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）  

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/edm/2023

Web of Science

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IFIP Advances in Information and Communication Technology  

Teaching and learning from one another is one of the most effective ways for learners to acquire proactive learning attitudes. In this study, we propose a new learning support system that encourages mutual teaching and learning by introducing a mechanism that guarantees sustainability. Learners submit articles called “learning-articles” that summarize their own learning and knowledge. The proposed system not only accumulates and publishes these articles but also has a mechanism to encourage the submission of necessary topics. The proposed system has been in operation since the academic year 2020, and it has collected learning-articles across our university’s nine academic disciplines from more than 300 learners. To investigate the effects of sharing learning-articles on education from the learners’ perspectives, a questionnaire was distributed among 25 students.

DOI： 10.1007/978-3-031-43393-1_44

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IFIP Advances in Information and Communication Technology  

Smart agriculture has assumed increasing importance due to the growing age of farmers and a shortage of farm leaders. In response, it is crucial to provide more opportunities to learn about smart agriculture at agricultural colleges and high schools, where new farmers are trained. In agricultural education, a system is used for managing environmental information, such as temperature and humidity, obtained from sensors installed in the field. However, it is difficult to make effective use of this system due to the time required to detect changes in the field interfering with class time and the problem of oversight. In this study, we proposed a field environment digest system that will help learners by providing the summarized field sensing information, and support them in analyzing the data. In addition, to examine the potential for using field sensing information in agricultural education, we investigated the usefulness of the summarized sensor information and students’ usage of this information. In this paper, we outline the contents of the developed system and the results of the digest evaluation experiments.

DOI： 10.1007/978-3-031-43393-1_10

Scopus

researchmap

掲載種別：研究論文（国際会議プロシーディングス）  

In recent years, several kinds of e-learning systems, such as e-book and Learning Management System (LMS) have been widely used in the field of education. When students access these systems, their activities on the systems will be continuously and automatically recorded and stored as learning logs. As the learning logs are stored in association with students and indicate students' learning activities, most studies have been “student-based” learning log analyses focused on students and each student's learning behavior. However, the “student-based” learning log analysis focuses on each student's learning behavior during the entire lesson (for example, studied well or didn't study enough) and cannot show what they learned. Therefore, if there is a need to investigate students' learning behavior regarding each topic of the lesson, such as which topic is learned well and which not in order to optimize the syllabus, we cannot conduct “student-based” learning log analysis directly. Instead of “student-based” learning log analyses, this study describes a method of “learning-topic-based” learning log analysis. We will show how to convert a learning log associated with students into a learning-topic-associated one and shape the logs into a two-dimensional matrix of learning topics and learning activities. Then we apply Non-negative Matrix Factorization (NMF) to the matrix in order to extract the learning patterns by activity. In addition, we make a three-dimensional matrix (tensor) of students, learning topics, and learning activities by subdividing the learning activities of each learning topic by students. We then apply Non-negative Tensor Factorization (NTF) to the tensor to extract detailed learning patterns. The methods proposed in this study will help teachers to have a comprehensively view of students' learning behaviors towards each learning topic easily even if the learning log is in a large-scale, so teachers can adjust syllabus according to the attracted learning behaviors, which is helpful to increase learning efficiency.

Scopus

researchmap

出版者・発行元：30th International Conference on Computers in Education Conference, ICCE 2022 - Proceedings  

In recent years, several kinds of e-learning systems, such as e-book and Learning Management System (LMS) have been widely used in the field of education. When students access these systems, their activities on the systems will be continuously and automatically recorded and stored as learning logs. As the learning logs are stored in association with students and indicate students' learning activities, most studies have been “student-based” learning log analyses focused on students and each student's learning behavior. However, the “student-based” learning log analysis focuses on each student's learning behavior during the entire lesson (for example, studied well or didn't study enough) and cannot show what they learned. Therefore, if there is a need to investigate students' learning behavior regarding each topic of the lesson, such as which topic is learned well and which not in order to optimize the syllabus, we cannot conduct “student-based” learning log analysis directly. Instead of “student-based” learning log analyses, this study describes a method of “learning-topic-based” learning log analysis. We will show how to convert a learning log associated with students into a learning-topic-associated one and shape the logs into a two-dimensional matrix of learning topics and learning activities. Then we apply Non-negative Matrix Factorization (NMF) to the matrix in order to extract the learning patterns by activity. In addition, we make a three-dimensional matrix (tensor) of students, learning topics, and learning activities by subdividing the learning activities of each learning topic by students. We then apply Non-negative Tensor Factorization (NTF) to the tensor to extract detailed learning patterns. The methods proposed in this study will help teachers to have a comprehensively view of students' learning behaviors towards each learning topic easily even if the learning log is in a large-scale, so teachers can adjust syllabus according to the attracted learning behaviors, which is helpful to increase learning efficiency.

Scopus

掲載種別：研究論文（国際会議プロシーディングス）  

Since the demand for programmers is increasing, programming courses are being offered widely. In this context, students' motivation can be damaged by difficulties they encounter in their programming courses. Although teachers' support is necessary to prevent such an issue, it is impossible for teachers to directly monitor all students' programming activities at the same time and determine which students have troubles with programming. Therefore, several studies have been conducted to help teachers monitor students. However, these studies do not provide an understanding of the activities of students who do not run their code, which may lead researchers to miss students who are in trouble. In this paper, we propose an indicator for detecting students who need coding support by analyzing programming logs that are recorded even when the students do not run their code. This gives teachers deeper insight into the students' programming performance. Although further work remains, the validation of this indicator shows that it could detect those students who are in trouble.

Scopus

researchmap

出版者・発行元：30th International Conference on Computers in Education Conference, ICCE 2022 - Proceedings  

The use of e-book reading systems such as Bookroll, and their ability to record readers' activities allows the design of predictive models capable of identifying at-risk students from their reading characteristics. Even though previous works have obtained promising results in this task, these results may not evidence the expected prediction performance in practical applications due to their selected assessment methods. Accordingly, in this paper, we assess this performance in two practical scenarios. The first is when we keep stored data from previous years of our course which can be used to train our model, and the second is when we only have data from a different course to use in this training process. In order to obtain a more accurate assessment, we collected 92, 574 samples of predictive performances from different models under the above-mentioned conditions. We also considered different feature representations along with variational latent representations, which can leverage our previous data to automatically design general hidden features. From our results, we understand that in the first condition we can expect a relatively good predictive performance, especially when using variational latent representations. However, in the second condition we found that even when using them, the predictive performances are very limited resulting in an impractical solution.

Scopus

掲載種別：研究論文（国際会議プロシーディングス）  

The use of e-book reading systems such as Bookroll, and their ability to record readers' activities allows the design of predictive models capable of identifying at-risk students from their reading characteristics. Even though previous works have obtained promising results in this task, these results may not evidence the expected prediction performance in practical applications due to their selected assessment methods. Accordingly, in this paper, we assess this performance in two practical scenarios. The first is when we keep stored data from previous years of our course which can be used to train our model, and the second is when we only have data from a different course to use in this training process. In order to obtain a more accurate assessment, we collected 92, 574 samples of predictive performances from different models under the above-mentioned conditions. We also considered different feature representations along with variational latent representations, which can leverage our previous data to automatically design general hidden features. From our results, we understand that in the first condition we can expect a relatively good predictive performance, especially when using variational latent representations. However, in the second condition we found that even when using them, the predictive performances are very limited resulting in an impractical solution.

Scopus

researchmap

出版者・発行元：30th International Conference on Computers in Education Conference, ICCE 2022 - Proceedings  

Since the demand for programmers is increasing, programming courses are being offered widely. In this context, students' motivation can be damaged by difficulties they encounter in their programming courses. Although teachers' support is necessary to prevent such an issue, it is impossible for teachers to directly monitor all students' programming activities at the same time and determine which students have troubles with programming. Therefore, several studies have been conducted to help teachers monitor students. However, these studies do not provide an understanding of the activities of students who do not run their code, which may lead researchers to miss students who are in trouble. In this paper, we propose an indicator for detecting students who need coding support by analyzing programming logs that are recorded even when the students do not run their code. This gives teachers deeper insight into the students' programming performance. Although further work remains, the validation of this indicator shows that it could detect those students who are in trouble.

Scopus

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（学術雑誌）   出版者・発行元：Sustainability (Switzerland)  

The COVID-19 pandemic has led to an increase in online classes, and programming classes are no exception. In such a learning environment, understanding every student’s programming process is mostly impractical for teachers, despite its significance in supporting students. Giving teachers feedback on programming processes is a typical approach to the problem. However, few studies have focused on visual representations of the evolution process of source-code contents; it remains unclear what visual representation would be effective to this end and how teachers value such feedback. We propose two feedback tools for teachers. These tools visualize the temporal evolution of source-code contents at different granularities. An experiment was conducted in which several university teachers performed a user evaluation of the tools, particularly with regard to their usefulness for reviewing past programming classes taught by another teacher. Questionnaire results showed that these tools are helpful for understanding programming processes. The tools were also found to be complementary, with different aspects being highly evaluated. We successfully presented concrete visual representations of programming processes as well as their relative strengths and weaknesses for reviewing classes; this contribution may serve as a basis for future real-time use of these tools in class.

DOI： 10.3390/su14138084

Web of Science

Scopus

researchmap

記述言語：その他   掲載種別：研究論文（学術雑誌）   出版者・発行元：Sustainability (Switzerland)  

While positive effects of imitating other learners have been reported, the recent increases in the number of online classes have seriously limited opportunities to learn how others are learning. Providing information about others’ learning activities through dashboards could be a solution, but few studies have targeted learning activities on e-textbook systems; it remains unclear what information representations would be useful and how they would affect learning. We developed a dashboard system that enables live sharing of students’ learning activities on e-textbooks. An experiment was conducted applying the dashboard in an online class to evaluate its impact. The results of questionnaires and quizzes were analyzed along with learning activities on the e-textbook system. From the questionnaire results, the most useful feedback types were identified. Regarding the impact on learning, the study found that a higher percentage of students who used the dashboard followed the progress of the class than those who did not. The study also found that students who used the dashboard were more likely to achieve higher quiz scores than those who did not. This study is the first to reveal what specific feedback is useful and to successfully investigate the impact of its use on learning.

DOI： 10.3390/su14126946

Web of Science

Scopus

researchmap

記述言語：その他   掲載種別：研究論文（学術雑誌）   出版者・発行元：Theoretical Computer Science  

The authors regret that the proof for Theorem 4 in the article “On the computing powers of [Formula presented]-reductions of insertion languages” was incomplete, which was pointed out by Kaoru Fujioka, Fukuoka Women's University, Japan. The proof needs some supplementary corrections for constructing the insertion system γ and [Formula presented]. Specifically, in the proof (on page 231), 1. line 4: Correct as [Formula presented], we construct the following rules: [Formula presented] 3. after line 11: Insert the following: • If there exists a rule

DOI： 10.1016/j.tcs.2021.06.028

Web of Science

Scopus

researchmap

記述言語：その他   掲載種別：研究論文（学術雑誌）   出版者・発行元：Acta Informatica  

Let K and L be two languages over Σ and Γ (with Γ ⊂ Σ), respectively. Then, the L-reduction of K, denoted by K%L, is defined by {u0u1⋯un∈(Σ-Γ)∗∣u0v1u1⋯vnun∈K,vi∈L(1≤i≤n)}. This is extended to language classes as follows: K%L={K%L∣K∈K,L∈L}. In this paper, we investigate the computing powers of K%L in which K ranges among various classes of INSji and min-LIN, while L is taken as DYCK and F, where INSji: the class of insertion languages of weight (j, i), min-LIN: the class of minimal linear languages, DYCK: the class of Dyck languages, and F: the class of finite languages. The obtained results include:INS11%DYCK=REINSi0%F=INSj1%F=CF (for i≥ 3 and j≥ 1)INS20%DYCK=INS20min-LIN%F1=LIN where RE, CF, LIN, F1 are classes of recursively enumerable, of context-free, of linear languages, and of singleton languages over unary alphabet, respectively. Further, we provide a very simple alternative proof for the known result min-LIN%DYCK2=RE. We also show that with a certain condition, for the class of context-sensitive languages CS, there exists no K such that K%DYCK=CS, which is in marked contrast to the characterization results mentioned above for other classes in Chomsky hierarchy. It should be remarked from the viewpoint of molecular computing theory that the notion of L-reduction is naturally motivated by a molecular biological functioning well-known as RNA splicing occurring in most eukaryotic genes.

DOI： 10.1007/s00236-022-00418-0

Web of Science

Scopus

researchmap

その他リンク： https://link.springer.com/article/10.1007/s00236-022-00418-0/fulltext.html

記述言語：その他  

Exploring the use of probabilistic latent representations to encode the students' reading characteristics

記述言語：その他  

The 4th Workshop on Predicting Performance Based on the Analysis of Reading Behavior

記述言語：その他  

記述言語：その他  

New Perspective on Input Feature Analysis for Early Feedback by Student Performance Prediction Considering the Future Effect

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）   出版者・発行元：ACM International Conference Proceeding Series  

Handwritten notes are one important component of students' learning process, which is used to record what they have learned in class or tease out knowledge after class for reflection and further strengthen the learning effect. It also helps a lot during review. We hope to divide handwritten notes (Japanese) into different parts, such as text, mathematical expressions, charts, etc., and quantify them to evaluate the condition of the notes and compare them among students. At the same time, data on students' learning behaviors in the course are collected through the online education platform, such as the use time of textbook and attendance, as well as the scores of the online quiz and course grade. In this paper, the analysis of the relationship between the segmentation results of handwritten notes and learning behavior are reported, as well as the research on automatic page segmentation based on deep learning.

DOI： 10.1145/3506860.3506915

Web of Science

Scopus

researchmap

その他リンク： https://dl.acm.org/doi/pdf/10.1145/3506860.3506915

記述言語：その他  

Coding Trajectory Map: Student Programming Situations Made Visually Locatable

記述言語：その他   掲載種別：研究論文（学術雑誌）   出版者・発行元：New Generation Computing  

We propose a new type of computing devices based on grammatical formulation augmented by multiset storages, called chemical reaction regular grammars (CRRGs), and investigate some formal language theoretic characterizations of CRRGs including their generative capabilities. Shortly, a CRRG is a regular grammar with multisets, while its computational capability exhibits very intriguing aspects depending on the manners of rule applications. Firstly, we show that the class of languages (denoted by CRRLλ) generated by CRRGs coincides with the class of languages accepted by chemical reaction automata (Okubo and Yokomori in Nat Comput 15(2): 215–224, 2016), whose implication is that the computing power of CRRGs is also equivalent to that of several known devices introduced from different motivations such as Petri nets (Peterson in ACM Comput Surv 9(3):223–252, 1977) and partially blind 1-way multicounter machines (Greibach in Theor Comput Sci 7:311–324, 1979). Second, a new manner of rewriting strategy is integrated into CRRGs and we show that CRRGs working in maximal-sequential manner can generate any recursively enumerable language, which is an unexpected result with a surprise. In contrast, it is also shown that regulated controls due to regular sets and matrix constraints do not enhance the computing power of CRRGs. Third, for each k≥ 1 a subclass of languages k-CRRLλ is considered, where k is the number of different symbols for multisets of CRRGs. We show that the class of languages is a full principal semi-AFL, which is obtained from a characterization result that L is in k-CRRLλ iff L= h(g- 1(Bk) ∩ R) for some homomorphisms g, h, a regular set R, where Bk is a paritally balanced language over k-symbol alphabet.

DOI： 10.1007/s00354-022-00160-8

Web of Science

Scopus

researchmap

その他リンク： https://link.springer.com/article/10.1007/s00354-022-00160-8/fulltext.html

researchmap

researchmap

researchmap

researchmap

researchmap

Web of Science

出版者・発行元：Proceedings - 2022 IEEE International Conference on Teaching, Assessment and Learning for Engineering, TALE 2022  

Students' performance prediction is essential for instructors to observe each student's learning behavior to discover which students have become at-risk. Early prediction helps instructors to intervene in time and provide academic support to these students. However, instructors should grasp essential behavior points to survey students' academic performance. In this study, we propose the Scaled-Dot Product Attention that can mine the relationship between the student's learning behaviors and performance to find the essential features that directly affect students' performance. In this study, we tested the early prediction performance of Scaled-Dot Product Attention with conventional algorithms. We then investigated essential lectures or features related to students' learning activities. From the result, we found that Scaled-Dot Product Attention outperformed the conventional algorithms to identify at-risk students and found the important lectures and students' actions.

DOI： 10.1109/TALE54877.2022.00059

Scopus

出版者・発行元：CEUR Workshop Proceedings  

Student Performance Prediction is essential for learning analysis of the students' learning behavior to discovering at-risk students for the early invention to support students. This study transforms the students' reading behavior into a two-dimensional matrix input based on each lecture material's reading behavior. The matrix input will be updated by accumulating the value for each week for performance prediction week by week. The multilayer perceptron neural network is employed to receive the matrix input and give feedback as a student's criteria consist of at-risk or no-risk students. This study considers the accuracy of a model considering between on contents information and weekly information. We also investigate the switching of learning materials' order, the feature importance of the reading operation on an event stream, and the difference in reading behavior between at-risk and no-risk students. These can help the instructors for an early invention to support at-risk students.

Scopus

出版者・発行元：CEUR Workshop Proceedings  

The emergence of digital textbook reading systems such as Bookroll, and their ability of recording reader interactions has opened the possibility of analyzing the students reading behaviors and characteristics. To date, several works have conducted compelling analyses characterizing the different types of students with the use of clustering ML models, while others have used supervised ML models to predict their academic performance. The main characteristic these models share is that internally they simplify the students' data into a latent representation to get an insight or make a prediction. Nevertheless, these representations are oversimplified, otherwise difficult to interpret. Accordingly, the present work explores the use of Variational Autoencoders to make more interpretable and complex latent representations. After a brief description of these models, we present and discuss the results of four explorative studies when using the LAK22 Data Challenge Workshop datasets. Our results show that the probabilistic latent representations generated by the proposed models preserve the student reading characteristics, allowing a better visual interpretation when using 3 dimensions. Also, they allow supervised regressive and classification models to have a more stable and less overfitted learning process, which also allows some of them to make better score predictions.

Scopus

Web of Science

出版者・発行元：AVSS 2022 - 18th IEEE International Conference on Advanced Video and Signal-Based Surveillance  

Background subtraction networks outperform traditional hand-craft background subtraction methods. The main advantage of background subtraction networks is their ability to automatically learn background features for training scenes. When applying the trained network to new target scenes, adapting the network to the new scenes is crucial. However, few studies have focused on reusing multiple trained models for new target scenes. Considering background changes have several categories, such as illumination changes, a model trained for each background scene can work effectively for the target scene similar to the training scene. In this study, we propose a method to ensemble the module networks trained for each background scene. Experimental results show that the proposed method is significantly more accurate compared with the conventional methods in the target scene by tuning with only a few frames.

DOI： 10.1109/AVSS56176.2022.9959316

Web of Science

Scopus

Web of Science

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s41965-021-00070-6

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.tcs.2020.11.029

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s10758-020-09436-8

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1186/s41039-019-0119-y

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.tcs.2019.01.032

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.1007/978-3-030-22580-3_35

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1080/10447318.2018.1543077

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.1109/TALE.2018.8615254

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.1007/978-3-319-94418-0_42

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.1007/978-3-319-91131-1_6

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1109/TLT.2017.2682086

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

DOI： 10.1145/3170358.3170412

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

© 2018 CEUR-WS. All Rights Reserved. In this paper, we explore the factor for improving the performance of prediction of students’ quiz scores by using a Recurrent Neural Network. The proposed method is applied to the log data of 2693 students in 15 courses that were conducted with following the common syllabus by 10 teachers. The experimental results show that in the case where the same teacher is not included in both training and test data, the accuracy of prediction slightly lower. We also show that at the beginning of a course, it is better to construct a prediction model including various items of learning logs, however, in the latter half, it is better to update the model by using selected information only.

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.1186/s41039-017-0053-9

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：その他   掲載種別：研究論文（学術雑誌）  

DOI： 10.3233/FI-2017-1569

記述言語：その他  

DOI： 10.1007/978-3-319-55345-0_13

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：その他   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：日本語   掲載種別：研究論文（大学，研究機関等紀要）  

記述言語：日本語   掲載種別：研究論文（大学，研究機関等紀要）  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語   掲載種別：研究論文（学術雑誌）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1051/ita/2013047

記述言語：英語   掲載種別：研究論文（大学，研究機関等紀要）  

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.tcs.2012.03.024

記述言語：英語   掲載種別：研究論文（大学，研究機関等紀要）  

開催年月日： 2016年5月

記述言語：英語  

国名：グレートブリテン・北アイルランド連合王国(英国)  

開催年月日： 2016年5月

記述言語：英語  

国名：グレートブリテン・北アイルランド連合王国(英国)  

開催年月日： 2016年5月

記述言語：英語  

国名：グレートブリテン・北アイルランド連合王国(英国)  

開催年月日： 2016年5月

記述言語：英語  

国名：グレートブリテン・北アイルランド連合王国(英国)  

開催年月日： 2016年5月

記述言語：英語  

国名：グレートブリテン・北アイルランド連合王国(英国)  

記述言語：日本語  

国名：日本国  

記述言語：日本語   会議種別：口頭発表（一般）  

国名：日本国  

記述言語：日本語   会議種別：口頭発表（一般）  

開催地：早稲田大学   国名：日本国  

記述言語：日本語   会議種別：口頭発表（一般）  

開催地：奈良教育大学   国名：日本国  

記述言語：英語  

国名：日本国  

記述言語：英語  

国名：台湾  

記述言語：英語  

国名：台湾  

記述言語：日本語  

国名：日本国  

記述言語：英語  

国名：中華人民共和国  

記述言語：英語  

国名：中華人民共和国  

記述言語：英語  

国名：中華人民共和国  

記述言語：英語  

国名：中華人民共和国  

記述言語：英語  

国名：中華人民共和国  

記述言語：英語  

国名：中華人民共和国  

記述言語：英語  

国名：中華人民共和国  

記述言語：英語  

国名：中華人民共和国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：英語  

国名：インド  

記述言語：英語  

国名：カナダ  

記述言語：英語  

国名：カナダ  

記述言語：英語  

国名：カナダ  

記述言語：英語  

国名：カナダ  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

記述言語：日本語  

国名：日本国  

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

出版者・発行元：CEUR Workshop Proceedings  

Scopus

J-GLOBAL

researchmap

J-GLOBAL

researchmap

記述言語：日本語  

特集「ラーニング・アナリティクス最前線」

DOI： 10.18910/70428

種別：大会・シンポジウム等 

種別：査読等 

外国語雑誌　査読論文数：3

国際会議録　査読論文数：5

種別：大会・シンポジウム等 

種別：査読等 

外国語雑誌　査読論文数：6

国際会議録　査読論文数：1

種別：査読等 

外国語雑誌　査読論文数：1

種別：大会・シンポジウム等 

種別：大会・シンポジウム等