九州大学 研究者情報
論文一覧
伊東 栄典(いとう えいすけ) データ更新日:2024.03.18

准教授 /  情報基盤研究開発センター 応用データ科学研究部門


原著論文
1. 松田 真治,伊東 栄典, 教師あり学習と強化学習を用いた麻雀AIの開発, 人工知能学会, 第124回知識ベースシステム研究会 (SIG-KBS), 124, 28-33, 2020.11, [URL], In recent years, game AI has been remarkably evolved. Some game AIs have outperformed top human players for complete information games such as Chess, Shogi, and Go. Compared with AI for perfect information games, game AI is not so strong for imperfect information games such as Poker and Mahjong. We research and development Mahjong AI using machine learning method. In this paper, we divide the internal Mahjong AI function into two parts: supervised learning and reinforcement learning. In the supervised learning part, the AI learns the choices of tiles using the top ranked human players in Tenho's game records. In the second part, the AI is reinforced selection function during many Mahjong games against three AIs created in the first step supervised learning..
2. 堺 雄之介, 伊東 栄典, 動画サイトにおける視聴者コメントの特徴抽出, 人工知能学会 第124回知識ベースシステム研究会 (SIG-KBS), 124, 17-22, 2021.11, [URL], In recent years, slander and bullying have become serious problems on SNS and video sharing service such as YouTube. Our research purpose is detection of comment flaming and extraction features using machine learning. We focus on YouTube, because YouTube is the largest video sharing site, massive viewers are watching movies everyday, and a lot of comments are posted. In this paper, we report comment collection method, flame judgment of comments, vectorization, and flaming detection. Additionally, we also report feature extraction method using
machine learning..
3. 堺 雄之介, 竹内 幹太, 伊東 栄典, コメントを利用した炎上動画検出に関する検討, 情報処理学会 研究報告 知能システム(ICS), 2021-ICS-203, 9, 1-5, 2021.07, [URL].
4. 松田 真治,伊東 栄典, 麻雀における他家の待ち牌推測, 情報処理学会 研究報告 数理モデル化と問題解決(MPS), 2020-MPS-130, 4, 1-6, 2020.09, [URL], 本研究の目的は、NN を用いて麻雀における他家の待ち牌の予測をすることである。 先行研究において、相手 の状態を推測する場合、時系列のない情報を利用していた。そこで本研究では、牌の切られた順序を入力として学習 を行った。その結果、先行研究と比較して、4.16 % f1score を向上させることができた。.
5. 堺 雄之介,伊東 栄典, 単語の分散表現に基づく流行語分析, 情報処理学会 研究報告 数理モデル化と問題解決(MPS), 2020-MPS-130, 6, 1-8, 2020.09, [URL], Twitter 等の SNS や,利用者が動画や小説等を投稿する CGM では消費者の生の声が反映される. そのため,消費者の嗜好や流行の分析対象に適している.流行語の抽出手法として単純に文書中の単語出 現頻度を用いる手法や,流行語抽出対象とする期間と,その前の期間での出現頻度の差分を取る手法があ る.本研究では流行語抽出の新たなアプローチとして,単語の類似語に注目した手法を提案する.また提 案手法を可視化する流行語分析ツールを作成した.実際に「syosetu.com」の小説メタデータ群に,提案手 法とツールを適用した.本論文では,提案手法,ツール,および適用結果について述べる..
6. 佐嘉田 悠樹,伊東 栄典, 間取り図に基づく類似性抽出, 第12回データ工学と情報マネジメントに関するフォーラム(DEIM2020), A6-2, 2020.03, 近年,FinTech のように IT 導入が各業界へ波及している.不動産業界でも「不動産テック (Real Estate Tech: ReTech)」が 注目されている.日本の不動産業界は,取引の不透明性や複雑な制度によりデジタル化が妨げら れていたものの,法改正や不動産ベンチャーの参入増大など徐々に緩和の動きがある.不動産テックの普及により,情 報量の増大,取引の活性化,ビッグ データ活用の実現,さらに潜在需要の掘り起こしの可能性がでている. 現在,日 本の不動産情報サイトは各不動産管理会社入力の文字情報を検索できる.しかし画像情報は検索できない.入力欄が 有る文字列情報と異なり,間取り図は不動産管理会社独自の形式で作成されており互換性が無い.本稿では,間取り 図を検索する手法を取り扱う.本研究では検索クエリとして利用者が間取りを想定して組み立てたブロック模型の写 真を与え,類似する間取り図画像を表示する手法を考える.Deep CNN モデルの VGG16 を用いて, 間取り図画像と クエリ画像を転移学習させる.続いて,VGG16 から特徴量を抽出し,CNN でクエリ画像と間取り図画像の一致・不 一致を学習させて,Siamese Network を構成する.最後に,間取り図と模型写真の類似度を算出する..
7. 弓場 邦哲, 伊東 栄典, 感情Tagを用いた感情学習及びLSTM,GRUの比較実験 ~デスクトップマスコットのAIエージェント化に向けて~, 電子情報通信学会 信学技報 AI2019-35, 119, 317, 31-36, 2019.11, [URL], 対話プログラムは、GRUやサーバーによる学習の普及により、大きくその精度を上昇させた。しかし、その個性は限られており、ユーザー毎に好みのプログラムを提供するサービスは存在しない。そこで、ユーザーごとに好みのキャラクターで作成されていたデスクトップマスコットに対話技術を組み込むことで解決を図ることにした。その手始めとして、感情学習と対話技術で多用されるLSTMとGRUの比較実験を行った。.
8. 柴田 知親, 伊東 栄典, 商品レビューの有用性ランキング推定, 人工知能学会 SIG-KBS, B5, 2, 25-30, 2019.11, On a real estate portal site in Japan, users can search for text information but not image information. This is because textual data entered by property management company is in an available format for searching, but image data is not. The floor plans which each property management company made in their respective formats are not compatible unlike the character provided in each fixed entry field. Therefore, the user needs to infer what kind of structure from the input character strings (number of rooms, facilities, area, etc.). In this study, we consider a system to search for similar floor plan images by giving a photograph of a construction toy structure as a search query in order to search for floor plan images. Deep CNN model (VGG16) is fine-tuned separately for the floor plan image and the query image. Subsequently, each feature vector is extracted from each learning layer of VGG16. Siamese Network is constructed to calculate similarity between the floor plan image and the query image by CNN learning the match / mismatch between the floor plan image and the query image. This paper gives an overview of the experimental procedures and results performed. This experiment deals with preprocessing of floor plan images and learning of floor plan images. In another experiment, the character information was extracted from the floor plan images..
9. 佐嘉田 悠樹, 伊東 栄典, 不動産間取り図画像の特徴量による物件類似性の導出, 人工知能学会 SIG-KBS, B5, 2, 19-24, 2019.11, [URL], On a real estate portal site in Japan, users can search for text information but not image information. This is because textual data entered by property management company is in an available format for searching, but image data is not. The floor plans which each property management company made in their respective formats are not compatible unlike the character provided in each fixed entry field. Therefore, the user needs to infer what kind of structure from the input character strings (number of rooms, facilities, area, etc.). In this study, we consider a system to search for similar floor plan images by giving a photograph of a construction toy structure as a search query in order to search for floor plan images. Deep CNN model (VGG16) is fine-tuned separately for the floor plan image and the query image. Subsequently, each feature vector is extracted from each learning layer of VGG16. Siamese Network is constructed to calculate similarity between the floor plan image and the query image by CNN learning the match / mismatch between the floor plan image and the query image. This paper gives an overview of the experimental procedures and results performed. This experiment deals with preprocessing of floor plan images and learning of floor plan images. In another experiment, the character information was extracted from the floor plan images..
10. 淀川 翼, 加藤 一成, 伊東 栄典, 単語の分散表現を用いた文書クラスタのラベル推定, 人工知能学会 SIG-SWO, 49, 3, 2019.11, [URL], Clustering is applied to divide customers into small subsets to analyze detail of customers. The attributes of each subsets are manually researched by human analysts. In case of documents, human analysts can extract the attributes of each subdocument set generated by clustering, if they check all documents. However, it is not practical to analyze by human if the size of documents is large. So, mechanical attributes estimation is required. In this paper, we propose a method for estimating the labels. The proposed method consists of three steps. At first, obtain distributed representation of words by fastText and large corpus. Next, extract feature words using SVM discriminator. Finally, estimate appropriate labels of a document set using similarity of word vector and SVM feature words. As an experiment, we apply proposed method to two document sets, The 20 newsgroups and Livedoor news corpus. Both sets are used for classification problem. We report the proposed method and experimental results..
11. Takuya Kawatani, Eisuke Itoh, Sachio Hirokawa, Tsunenori Mine, Location does not always determine sudden braking, 2019 IEEE Intelligent Transportation Systems Conference (ITSC), 10.1109/ITSC.2019.8917480, 27-30, 2019.11, [URL], Understanding conditions and situations causing sudden braking is important for preventing traffic accidents. Previous studies have used probe vehicle data to detect risky positions where sudden braking frequently occurred. However, they have mainly focused on vehicle-related factors. In this paper, we propose a novel method for discriminating sudden braking. Unlike previous studies, the method exhaustively explores probe data including temporal factors, constructs a large number of features combining pairs of feature names and their values, and applies the Support Vector Machine classifier and Feature Selection method to the features. To conduct the experiments, we used probe data provided by the Aizu-Wakamatsu City Open Data Utilization Verification Project. The proposed method discriminated sudden braking quite accurately, with a discrimination performance averaging an F1 measure of 93.2%. We also found that the probability of the occurrence of sudden braking is not always high at locations where sudden braking frequently occurred, but rather, temporal factors such as date and time, or day of week are strongly related to performance in discriminating sudden braking with high probability..
12. Takuya Kawatani, Eisuke Itoh, Sachio Hirokawa, Tsunenori Mine, Machine Learning and Visualization of Sudden Braking using Probe Data, 2019 8th International Congress on Advanced Applied Informatics (IIAI-AAI), 10.1109/IIAI-AAI.2019.00024, 67-72, 2019.07, [URL], This paper presents a novel mining and visualizing tool that detects features to estimate sudden braking. The tool uses a machine learning and feature selection method to find the features exhaustively from combinations of the features which include not only vehicle-related factors, but also outer circumstances or temporal factors. The tool also obtains the locations inferred by the features detected. A normal way would first search for locations where sudden braking behavior frequently occurred, but it is not always true that the occurrence probability of sudden braking at the locations is high. On the other hand, our tool finds the locations related to sudden braking with high probability, more than 98%. Through the visualizing process, the features can be used as clues to find new factors which affect sudden braking..
13. 川谷 卓哉, 廣川 佐千男, 峯 恒憲, 伊東 栄典,, 機械学習による走行プローブデータからの異常走行箇所発見, 情報処理学会 第76回高度交通システムとスマートコミュニティ研究会 (ITS76), 2019-ITS-76, 7, 1-6, 2019.02, [URL], 自動車の走行状況や道路状況などを知るため.スマートフォンセンサーをはじめ,GPS を利用した Automatic Vehicle Location (AVL) など,様々な装置を利用し,プローブデータ(自動車の走行データ)の獲得が行われて,機械学習手法を利用した走行挙動の分析が行われてきている.本研究では,機械学習手法と特徴選択手法を組み合わせた,通常とは異なる走行挙動(異常走行挙動)の識別手法を提案する.具体的には会津若松市オープンデータ活用実証事業により提供されている公用車・公共交通車両走行情報履歴データにSVM (Support Vector Machine) と特徴選出の手法を用いた異常走行挙動の識別手法を提案する. 本手法を適用した際の,異常走行挙動の識別結果と識別精度,識別に寄与した特徴についても考察する..
14. 飯田 委哉, 伊東 栄典, セレンディピティを考慮したCGM小説推薦, 人工知能学会 第15回データ指向構成マイニングとシミュレーション研究会, 2-0284, 2018.11, [URL], Recently, CGM (Consumer Generated Media) services become popular. Although
a huge amount of contents have been posted to CGM site, but only a few contents are selected
and viewed by users. This phenomenon are called as very short head and very long tails. Some
researcher mentioned that recommendation algorithms made influence to user’s content selection
strongly. Serendipitous recommendation is necessary for healthy cultural glowth of CGM contents.
In this paper, we propose a new serendipitious recommendation method for CGM novels. We apply clustering to novels to divide into middle size clusters, and contents are similar in the same cluster. For clustering, each content is represented into a vector using doc2vec. Next, we calculate the distance from the user’s preference (bookmark) to each clusters. We believe that contents in near but not nearest cluster are better for serendipitious recommendation. We apply our method to web CGM novels in syosetu.com, and we also construct a web based recommendation system..
15. 柴田 知親, 伊東 栄典, 商品レビュー文の有用性推定に関する考察, 人工知能学会 第15回データ指向構成マイニングとシミュレーション研究会, 2-0286, 2018.11, [URL], Product reviews can be helpful for consumers to finalize their purchasing decisions. The advertisement contains sentences intended by the seller. Meanwhile, the product review seems to be an honest impression of the consumer, so it is easier to trust. However, there are some problems such as fake reviews and guerrilla marketing in recent years. Therefore, in many e-commerce sites, each reviews have“Helpful” and “Not Helpful” buttons for consumers to evaluate the helpfulness of the review. In this paper, we consider the estimation model of review helpfulness for the review
sentences of Rakuten Ichiba (rakuten.co.jp)..
16. Yoshiaki Kasahara, Takao Shimayoshi, Eisuke Ito, Naomi Fujimura, The Past, Current, and Future of our Email Services in Kyushu University, ACM, 10.1145/3235715.3235737, 103-106, 2018.11, [URL], In Kyushu University, Information Infrastructure Initiative provides email service for students and staff members. Email services for students and staff members were started separately. For students, an email service was started as Unix accounts of "Computer System for Education'' in 1995. On the other hand, an email service for staff members was started in 2009, and eventually the two mail services were merged into the current "Kyushu University Primary Mail Service'' in 2014. The designs of these mail systems were affected by various operational issues and political decisions at their times. We think that running an in-house mail system is becoming less feasible due to the initial/operational cost, security issues, and our dwindling budget. For the current system, the planned 5-year lifetime ends in this fiscal year. Therefore, we are forced to migrate to a cloud-based mail service. In this presentation, we want to share our past experiences and future plans about our university email services..
17. Kazuhisa Noguchi, Eisuke Ito, Holonomic Function of 2 Parameter Logistic Model Item Response Theory Parameter Estimation, 2018 10th International Conference On Education Technology And Computers (ICETC2018), J113, 2018.10, IRT (Item Response Theory) is a theory for scoring of tests, and it used for some test systems such as TOEFL. IRT's estimation is item parameter for each question, and examinee’s ability parameter. We propose a new method to estimate parameter of IRT using the Holonomic Gradient Method. When we use Holonomic Gradient Method, we check target function is Holonomic Function. Our target function is 2 Parameter Logistic Model IRT's Likelihood Function. In this paper, we show target IRT's Likelihood Function is Holonomic Function..
18. Kazuhisa Noguchi, Eisuke Ito, Holonomic Approach for Item Response Theory Parameter Estimation,, 2018 7th International Congress on Advanced Applied Informatics(AAI2018), 10.1109/IIAI-AAI.2018.00068, 322-326, 2018.07, The IRT (item response theory) is a theory for scoring of tests, and it is used for some test systems such as TOEFL. Classical test scoring is the total of raw scores for each items (questions). On the other hand, IRT can make fine performance finer than the raw score method. IRT is based on the relationship between individual test takers’ performances on a test item (question), and the test takers’ levels of performance. Even with the same number of correct answers, the IRT score may be different depending on the degree of difficulty of item (question). Computational complexity of parameter estimation of IRT is one of important issue. To solve this issue, we propose a parameter estimation method of IRT using the Holonomic approach. Generate differential equations which solve the likelihood function of IRT. By solving this differential equation, it is able to estimate the ability parameter..
19. Eisuke Ito, Yuya Honda, Sachio Hirokawa, Empathy factor mining from reader comments of e-manga, Proc. of eKnow2018 (The Tenth International Conference on Information, Process, and Knowledge Management), 107-112, 2018.03, [URL], The digitization of manga (Japanese comics) is currently progressing. In the past, expressions of manga have evolved in a way that is suitable for printing on paper. With the spread of smartphones, the expression of e-manga is developing at present. In this research, we focus on reader comments on e-mangas in Comico. Comico is a popular e-manga service. Similar to other online contents services, such as YouTube, Comico implements a user comments system. Readers can post comments easily, and the comments quickly reach others, including the manga creator. Comico recognizes that reader comments may influence the creator and the story of e-manga. In this research, we try to mine the empathy factor of readers for the story and the characters of e-manga. We collect reader comments and apply feature selection of SVM (Support Vector Machine) to mine empathy factors among readers..
20. 佐嘉田 悠樹, 伊東 栄典, クラスタリングによる利用者投稿動画の偏在性分析, 第10回データ工学と情報マネジメントに関するフォーラム (DEIM2018), G4-2, 2018.03, 動画や小説,画像などのコンテンツを投稿するサービス(CGM, Consumer Generated Media) が人気である。近年CGMサイトへの投稿コンテンツの画一化への懸念がある。多様性が減少し画一化が進むと文化的な活力も減り,サイト経営にも問題になる。我々はニコニコ動画を対象に,動画の多様性動向について分析している。今回,動画のメタデータにクラスタリングを適用し,クラスタのサイズ等から動画集合の多様性について分析した。動画をクラスタリングするには,動画をベクトルで表現する必要がある。動画にはタイトル,説明文,視聴者が付与するタグが付随しているものの,これらの文字は重要情報ではあるものの,情報量が少なくまた単語のゆらぎも大きい。そこで,辞書であるニコニコ大百科等を利用して,動画ベクトル化する。ニコニコ大百科の説明ページから,Doc2Vecで見出し語を,Word2Vecで各単語をベクトル化した。単語のベクトルデータを用いて,各動画をベクトルで表現した。次に動画ベクトルの集合を分割X-meansでクラスラリングした。その後,各クラスタのサイズなどを分析することで,投稿動画の多様性を分析した,ベクトル化,クラスタリングの結果,および多様性動向の分析結果について報告する。.
21. 柴田 知親, 伊東 栄典, 回帰分析によるCGM動画再生回数推定, 第10回データ工学と情報マネジメントに関するフォーラム (DEIM2018), C5-2, 2018.03, 近年,ネットワークおよびモバイル端末の普及に伴い,Youtubeやニコニコ動画といった,利用者投稿型動画共有サービスが人気である.動画以外にも,小説家になろう,pixiv,comicoなどに代表される,小説やイラスト,漫画などの投稿サイトも人気である.これらのサイトは,CGM(Consumer Generated Media)とも呼ばれる.我々は,CGMの一つであるニコニコ動画を対象に,動画のランキングや推薦を研究してきた.他のCGMと同様に,ニコニコ動画では極少数の動画は大量に再生されるのに対し,ほとんどの動画は再生回数が少ないというロングテールの分布をする.利用者が何らかの選択基準により動画選択した結果が再生回数に現れていると考えられる.そこで,CGMにおける利用者のコンテンツ選択モデルを確立したいと考えている.本研究では,コンテンツ選択モデル確立のために,動画の再生回数を回帰分析で推定する.最初に線形回帰分析で再生回数の推定した.その結果,多様な動画が含まれていると推定精度が低くなった.次にDeep Learningを適用し,Multi Layer Perceptronによる再生回数の非線型回帰分析を行った.対象の動画集合の多様性を減らすため,クラスタリングで動画集合を分割した.分割した動画集合について,Deep Learningで再生回数を推定した.これらの手法および結果について報告する..
22. 飯田 委哉, 伊東 栄典, 佐嘉田 悠樹, クラスタリングによるオンライン小説の多様性動向分析, 情報処理学会 火の国情報シンポジウム2018, B1-4, 2018.03, 近年,動画ではYouTubeやニコニコ動画が,小説では小説家になろうといった 利用者投稿型のCGM (Consumer Generated Media)サイトが人気である. CGMサイトには毎日多数のコンテンツが投稿されており,また膨大な利用者がコンテンツを閲覧している. 現在,CGMサイトへの投稿コンテンツの画一化が懸念されている.既に見たことのあるようなコンテンツや派生コンテンツの増加が感じられる. 我々は今回,「小説家になろう」を対象に,Cos類似度の総和を調べることで小説家になろうの画一化について定量的に評価をおこなった.次に,偏りのあるジャンルを特定するために,クラスタリングによるオンライン小説の多様性動向を調査する.各小説のあらすじをDoc2Vecを用いてベクトルで表現し,最後に,小説のベクトル群をX-meansでクラスタリングする. クラスタ数やクラスタの要素数を分析することで,投稿小説の多様性動向の分析を行う..
23. 本田 優也, 伊東 栄典, 廣川 佐千男, 読者コメント連動型漫画における人物の登場頻度分析, 人工知能学会, 02_1-02_6, 2017.11, 現在漫画コンテンツの電子化が進んでいる。従来漫画の表現は,紙で印刷するものに適して進化してきた。スマートフォンの普及により,電子漫画の表現も変化している。電子コンテンツの場合,利用者(読者,視聴者)と作者の間の交流が,簡単かつ短時間でやりとり可能になっている。読者のコメントから,読者間の交流が進む場合もあるし,作者の意図しない方向に,読者の興味が盛り上がる場合もある。電子漫画の1つに「コミコ(http://www.comico.jp/)」がある。コミコはスマートフォンに最適な漫画作りを進めている。また,読者コメントの積極的な利用も進めている。コミコの運営サイトによると,コメントが作者と読者との相互作用で漫画に影響を与えていると延べている。本研究では,電子漫画に対する読者コメントに着目する。また,特に漫画のキャラクタに着目し,読者がキャラクタに持つ印象を分析する。コメントの収集手法,コメントの統計解析,最後にキャラクタの関係推移について分析手法と結果を述べる。.
24. Eisuke Ito, Yuya Honda, Keyword diversity trend of consumer generated novels, Proceedings of ICESS2017 (The Third International Conference on Electronics and Software Science), 2017.08, Recent years, CGM (Consumer Generated Me- dia), such as YouTube and nicovideo.jp for movies, syosetu.com for novel stories, become very popu- lar. A lot of contents are posted to CGM sites ev- ery day, and also a large number of users are en- joying posted contents. At present, some articles mentioned decreasing diversity of contents. Some posted new content may be similar with previous posted contents. The authors are afraid that de- creasing diversity of contents causes less energetic cultural activity. In this paper, the authors pro- posed two quantitative metrics of contents diversity, and applied them to the contents in syosetu.com. They focused the keywords which are given to the novel by the novel author, and calculated entropy and similarity of keywords. As the results, they ob- served increase of similarity, and it shows decrease of diversity of contents..
25. Kazuhisa Noguchi, Tomoya Iida, Eisuke Ito, An analysis of CGM contents pageview using SIR Model and GBM, Proceedings of ICCTD2017 (2017 8th International Conference on Computer Technologies and Development), 10.1145/3093241.3093264, 19-21, 2017.03, In consumer generated media (CGM) site, such as YouTube and nicovideo, only few contents are viewed very much, but most contents are only viewed few times. Our research target CGM sites are nicovideo.jp and syosetu.com. Nicovideo.jp is a popular movie CGM site in Japan and syosetu.com is the largest novel CGM site in Japan. We already found that pageview distribution of contents in both CGM sites follow a lognormal distribution. In this paper, we consider user’s content selection model which will lead lognormal distribution. We apply Geometric Brownian Motion model into SIR model. SIR model is used for simulation of population transition process or epidemic process of infection disease. In this paper, we report the results of some simulation..
26. Yoshiaki Kasahara, Takuya Kawatani, Eisuke Ito, Koichi Simozono, An Analysis of Relationship between Storage Usage Distribution and Per-User Quota Value, ACM, 10.1145/2974927.2974936, 153-158, 2016.11, [URL], To prevent resource (especially storage) shortage, information systems such as storage services and email services usually impose an upper bound of resource consumption (quota) per user. In a conservative way, an administrator tends to set a quota value such as the storage capacity divided by the expected maximum number of users for safety and fairness, but it tends to leave large unused storage space, because the users' storage usage pattern shows a long-tailed distribution. In this paper, we analyzed storage usage distribution of some email services to approximate the distribution using a power-law distribution, and proposed a method to calculate an optimal quota value from a target size of storage consumption to increase storage utilization. We applied an optimal quota value we calculated to a real email service and analyzed the effect of quota change. Then, we analyzed actual distributions further to find a better model to approximate the distribution, and found that a log-normal distribution explained the distribution better than power-law. We also analyzed two other universities' email service to find similar distribution in these systems..
27. Kyohei Kamihata, Eisuke Ito, A quantitative contents diversity analysis on a consumer generated media site,, International Society of Artificial Life and Robotics, 436-440, 2016.01, Recent years, CGM (Consumer Generated Media) services, such as YouTube and nicovideo, are growing into social contents communication media. A lot of movies are posted to a CGM site, and many users are viewing them every day. Some web news articles mentioned that the diversity of movie contents might decrease in CGM site. The authors believe that contents diversity is necessary to keep CGM site activity, and for cultural sustainability. The authors try to measure diversity of movie contents of a CGM site quantitatively. They propose two quantitative metrics of contents diversity, and apply them to a CGM contents in nicovideo.jp..
28. 林 驍, 伊東 栄典, 廣川 佐千男, 大規模ECサイトの商品レビュー傾向分析, 情報処理学会, 2015-ICS-181, 7, 1-6, 2015.12, [URL], 消費者による商品レビューは,購入時判断に重要である。広告には売り手の意思が入るのに対し,レビューは購入した客の正直な感想であるため信用できる。ネット上の通信販売では情報提供として消費者による商品レビューを記載している。我々は,ネット上のECサイトにおける商品レビューからの知識発見について研究している。やらせやステマ,誹謗などのノイズを除去し,かつ消費者に有用な情報が無いレビューを除去したいと考えている。本論文で,楽天市場で販売される商品についてのレビュー文を対象に,レビュー投稿行為の不自然さや,レビュー文の特徴分析を行った。また,購入していない商品に対するレビュー文を調べることで,楽天市場における特殊な商品レビューの傾向を理解した。これらの結果について報告する。.
29. 上畑 恭平, 伊東 栄典, タグの類似度に着目した利用者投稿サイト動画の多様性分析, 電子情報通信学会, 115, 381, 83-88, 2015.12, 近年,YouTubeやニコニコ動画などの利用者投稿型動画共有サービスが人気である.これらのサイトはCGM (Consumer Generated Media) とも呼ばれる.現在CGMサイトは社会に影響を与えるメディアに成長している.CGMサイトに毎日多数の動画が投稿されており,また膨大な利用者が動画を閲覧している. 現在,CGMサイトに投稿されるコンテンツの画一化が指摘されている.以前見たことのあるような動画や,派生動画が増えているように感じられる.コンテンツの多様性が減少して画一化が進むと,文化的多様性が失われると思われる.我々は,CGMサイトであるニコニコ動画を対象に,動画コンテンツの多様性動向を分析している.今回,動画に付与されたタグについて,動画タグ間の類似度に着目した分析を行った.その結果を報告する..
30. 川谷 卓哉, 伊東 栄典, 笠原 義晃, 下園 幸一, クォータ変更に伴うストレージ使用量変化の分析, 情報処理学会, 6-13, 2015.11, [URL], 電子メールなどの情報システムは,組織活動を支える基盤となっている.組織向けの情報システムを運用する部門では,そのシステムの想定使用期間における使用資源量の見積もりが重要である.ストレージを適切に用意するためには,ストレージ使用量の傾向を把握する必要がある.電子メールを含む多くの情報システムで,利用者のストレージ使用量はロングテール型の分布になる.筆者らは,ストレージ容量に対する適切なクォータの設定について,ストレージ使用量の統計分布に基づく予測について研究している.以前の統計分析に基づき,筆者らが所属する九州大学の電子メールシステムのクォータを,冪分布近似に基づく推定による値に変更した.今回,クォータに変更による,各利用者のストレージ使用量の分布の変化と,システム全体でのストレージ使用量増加の傾向の変化を報告する.また,筆者らが提案した統計分析による予測が適切かどうかも検証する..
31. Yoshiaki Kasahara, Naomi Fujimura, Eisuke Ito, Masahiro Obana, Introduction of Unchanging Student User ID for Intra-Institutional Information Service, ACM, 10.1145/2815546.2815578, 141-144, 2015.11, [URL], In Kyushu University, a traditional "Student ID" based on student number assigned by Student Affairs Department had been used as the user ID of various IT services for a long time. There were some security and usability concerns using Student ID as a user ID. Since Student ID was used as the e-mail address of the student, it was easy to leak outside. Student ID is constructed based on a department code and a serial number, so guessing other ID strings from one ID is easy. Student ID is issued at the day of the entrance ceremony, so it is not usable for pre-entrance education. Student ID will change when the student moves to another department or proceeds from undergraduate to graduate school, so he/she loses personal data when Student ID changes. To solve these problems, Kyushu University decided to introduce another unchanging user ID independent from Student ID. This paper reports the design of new user ID, ID management system we are using, and the effect of introduction of new user ID..
32. Yoshiaki Kasahara, Takuya Kawatani, Eisuke Ito, Koichi Simozono, Naomi Fujimura, Optimization of storage quota based on user’s usage distribution, IEEE, 10.1109/COMPSAC.2015.221, 149-154, 2015.07, To prevent shortage of storage space in a service system, an administrator usually set per-user quota as an upper limit of usable space for each user. To avoid service failure caused by resource exhaustion, the administrator tends to set a conservative quota value such as the storage capacity divided by the expected maximum number of users. In this research, we analyzed long-term storage usage history of our email system and file sharing system in Kyushu University. Mostly through the analyzed period, the usage pattern showed a long-tailed distribution similar to log-normal distribution. Also the overall storage consumption slowly increased during the analyzed period. Based on these analysis, we defined “storage utilization ratio” to evaluate how the storage was effectively used. By approximating a storage utilization pattern as a power-law distribution, we proposed a method to calculate the optimal quota value to maximize the utilization ratio..
33. Kensuke Baba, Toshiro Minami, Eisuke Ito, Modeling Changes in Demands for Books with Elapsed Time from Publication, Computational Science and Its Applications -- ICCSA 2015, 10.1007/978-3-319-21407-8, 9156, 172-181, 2015.06, Book selection in libraries should be conducted on the basis of analyses on circulation data. In addition to the number of loans, the change of the number over time can be a criterion for book selection. This paper proposes a model to represent the change of the number with the time from publication. The proposed model is applied to practical circulation data in a library and is evaluated in terms of the root mean square errors. As a result, the proposed model is suitable compared with the standard one. Additionally, this paper analyzes the differences of the changes between classified books which are obtained by the previous analysis. The information obtained by applying the proposed model to circulation data is expected to be utilized for book selection..
34. 川谷 卓也, 伊東 栄典, 笠原 義晃, 藤村 直美, 適切なクォータ値設定のためのディスク使用量分析, 情報処理学会, 2015-IOT-28, 32, 1-7, 2015.02, ストレージの容量不足を防ぐため,システム管理者は各利用者の使用上限値(クォータ値)を設定する.システム管理者はサービスの安定提供のために,少なめのクォータ値を設定しがちである.また,クォータ値に単純に容量を利用者数で均等割りした値を設定することも多い.本研究では,筆者らが所属する九州大学の電子メールとファイル共有(蓄積)システムを対象に,長期間のディスク使用量を分析した.ほとんどの時点で,利用者のディスク使用量はロングテール型の分布をしており,その分布は対数正規分布に近いことが分かった.長期間のディスク使用量の推移を調べた所,全体の使用量が微増していたことも分かった.ストレージの活用度を定義し,それに基づく評価方法を考案した.さらに,ディスク使用量分布の冪分布近似により,活用度が高くなるクォータ値の算定方法を考案した.実際のデータに基づき,ストレージの活用度を最高にするクォータ値も算出した..
35. Yoshiaki Kasahara, Eisuke Ito, Naomi Fujimura, Introduction of New Kyushu University Primary Mail Service for Staff Members and Students, ACM, 10.1145/2661172.2662965, 103-106, 2014.11, In the end of fiscal year 2013 (March 2014), Kyushu University Information Infrastructure Initiative introduced new Primary Mail Service for Staff Members and Students. The previous service for staff members had been built using proprietary mail and LDAP appliance, but tight user license and the nature of proprietary system caused several troubles. The previous service for students had been built using open source software (Postfix and Dovecot), but there were some issues with the old implementation of LDAP authentication. With these experiences, we decided to design a new system by merging both system using open source software including Postfix, Dovecot, OpenLDAP, and Squirrelmail. We also extended alias address setting service (previously available for students only) to all the members including staff members. In this paper, we explain the design and implementation, user migration, current status and future works of our new mail service..
36. Zechen Wu, Eisuke Ito, Correlation Analysis between User's Emotional Comments and Popularity Measures, Proceedings of IIAI-AAI 2014/ESKM 2014, 10.1109/IIAI-AAI.2014.65, 280-283, 2014.09, On user generated contents sites, user's comments include the user's impression to the content. The authors believe that user's comments can be used as the data mining resource to evaluate the contents. In this paper, the authors focus on bilibili.tv, a Chinese video sharing site, and analyze user's emotional comments on the site. Toward development of content quality measurement based on emotional comments, they investigate co-relation between emotional comments and popularity measures such as number of replay and bookmark. They pick up the phrase "233", which is a Chinese network jargon which represents lot of laugh, and studied the number of "233" related comments and other popularity measures..
37. 藤村 直美, 笠原 義晃, 伊東 栄典, 尾花昌浩, 井上 仁, 学生番号と異なる学内情報サービス専用ID付与, 情報処理学会, 2014-IOT-26, 1, 1-6, 2014.06, くの組織で,組織内情報サービスのための利用者認証基盤構築されている.九州大学でも統合認証基盤を構築しており,そのアカウントとして学生には学生番号に基づく利用者IDを発行してきた.学生番号に基づく利用者IDにはセキュリティの問題がある.学生用の基本メールアドレスに用いてきたため外部に漏れやすく,また学生番号は学部または大学院で連続した数値等を用いるため利用
者IDが推定されやすい.実際に学生アカウントを不正利用されるインシデントも発生した.学生番号に基づくIDは,入学前に情報サービスを利用出来ない,大学院進学時にアカウントが不連続になるなどの問題があった.これらの問題を解決するため,学内情報サービス専用のIDを付与する事にした.本稿では,利用者ID体系,利用者IDデータベース,認証用LDAPの構成などについて報告する..
38. Xiao Lin, Eisuke Ito, Sachio Hirokawa, Chinese Tag analysis for foreign movie contents, 13th IEEE/ACIS International Conference on Computer and Information Science (ICIS2014), 163-166, 2014.06, 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. .
39. Jun Zeng, Brendan Flanagan, Sachio Hirokawa, Eisuke Ito, A Web Page Segmentation Approach Using Visual Semantics, The Institute of Electronics, Information and Communication Engineers (IEICE), E97-D, 2, 223-230, 2014.02, [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..
40. Takuya Kawatani, Eisuke Ito, Kensuke Baba, A prototype of real-time seat usage analysis using smart sensors, IIAI AIT 2013 (Advanced Information Technologies 2013), P32.pdf, 2013.11, The authors consider analyzing real time seats usage of library browsing rooms using smart sensor and cloud computing infrastructure as a library marketing activity. The smart sensor box checks human existence periodically, and sends the sensed data to the sensor cloud through the Internet. We apply information retrieval technology to the accumulated data. This paper shows the research plan of real time seats usage ratio using smart sensor, and roughly shows cloud system architecture and application images..
41. Eisuke Ito, Sachio Hirokawa, Keyword relation analysis using concept graph toward automatic categorization of online novels, IIAI AIT 2013 (Advanced Information Technologies 2013), P24.pdf, P24.pdf, 2013.11, Contents of user-generated media (UGM) such as YouTube are popular in recent years. The authors are interested in online novel services as a UGM on the Web. A large number of novels are being uploaded, and a few novels become major. Because there are no professional editor or no trained librarian concerning to UGM, neither quality evaluation nor categorization of UGM contents is clear. Moreover, the creator of a novel gives keywords without control, and they often generate new words, then it is difficult to use keyword for categorization. Toward automatic categorization, we have to understand relation between keywords. In this paper, we focused on analysis of keywords relation using concept graph. The concept graph is based on statistical analysis of word frequency and co-occurrence. We also show several interesting relations on concept graph..
42. Eisuke Ito, Brendan Flanagan, CHENGJIU YIN, Tetsuya Nakatoh, Sachio Hirokawa, A Private Cloud Environment for Teaching Search Engine Construction, Proc. of ICCE2013 (The 21st International Conference on Computers in Education), 391-397, 2013.11, [URL], 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..
43. Eisuke Ito, Yoshiaki Kasahara, Naomi Fujimura, Implementation and operation of the Kyushu university authentication system, Proc. of ACM SIGUCCS 2013, 10.1145/2504776.2504788, 137-142, 2013.11, [URL], Nowadays, a university needs to build and maintain a central ID database and authentication system for better ICT (information and communication technology) services. In 2008, the headquarters of Kyushu University had defined medium-range policy of ICT infrastructure preparation, and the policy had indicated construction of a central authentication system. According to the policy, the authors elaborated an installation plan of the Kyu(Q)shu University authentication system (QUAS, for short). Since 2009, Information Infrastructure Initiative of Kyushu University, to which the authors belong, has been issuing ID cards to all employees, and also operating LDAP servers. This paper introduces the action plan and outline of QUAS. This paper also describes two recent topics of QUAS. One is high load of LDAP servers because of rapid increase of mobile devices, and the other one is development of a multifactor authentication Shibboleth Identity Provider (IdP)..
44. Yoshiaki Kasahara, Eisuke Ito, Naomi Fujimura, Gulliver's Toss: Google's Chronic Big Load to University Mail Server and Its Sudden Resolution, Proc. of ACM SIGUCCS 2013, 10.1145/2504776.2504815, 169-174, 2013.11, [URL], Traditionally, Kyushu University has been providing email service internally using its own domain name for staff members and students of the university. Around January 2012, we noticed that the high load of the university authentication server, and we realized that one of causes was the access from the mail server for students (called Student Primary Mail Service). Detailed analysis showed that there was chronic big load produced by Gmail's Mail Fetcher, especially toward nonexistent accounts removed due to graduation. In this paper, we explain the situation and reasons of the big load induced by Google, its possible countermeasures, and its sudden resolution by Google's silent change..
45. Yoshiaki Kasahara, Eisuke Ito, A study of network issues for implementing large-scale academic cloud service, IEICE (The Institute of Electrics, Information and Communication Engieers), 113, 240, 35-40, IA2013-32, 2013.10, [URL], The virtualization technology of computer resources has matured enough for production services, and cloud services became popular in various fields. Cloud services are also gradually incorporated in universities' research and education activities. There is a demand for cloud infrastructure suitable for academic activities. To implement practical and user-friendly services on private, community, or public cloud services, we need to consider about network design. This paper tries to clarify network issues for implementing cloud services considering use cases, the number of instances, utilization of IP addresses, virtual private network, client access networks, and so on..
46. Tetsuya Oishi, Masao Mori, Toshie Tanaka, Eisuke Ito, Cooperative Development and Individual Operation for Information Systems of IR in a Virtual Environment, AAI2013 (International Conference on Advanced Applied Informatics), 10.1109/IIAI-AAI.2013.68, 235-238, 2013.08, We propose a new method of cooperative development of information systems utilizing virtual machines. This method not only realizes individual operations of systems in each institution, but also is efficient for the institution in shortage of IT-talented staff..
47. Eisuke Ito, Takahiro Urakawa, Brendan Flanagan, Sachio Hirokawa, Keywords frequency trend analysis of online novels, AAI2013 (International Conference on Advanced Applied Informatics), 10.1109/IIAI-AAI.2013.92, 68-73, 2013.08, 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..
48. 笠原 義晃, 伊東 栄典, IaaSクラウド型教育情報システムの実現可能性調査, 情報処理学会 情処研報告, 2013-IOT-22, 14, 1-6, 2013.08, [URL], 仮想化技術の普及により,IaaS 型のクラウドサービスが普及している.IaaS クラウド基盤を使うことにより,大学での教育情報システムを外部へアウトソースすることが可能である.アウトソースできれば,維持コストの削減が可能である.我々は,現在の PC 教室型教育情報システムの,IaaS クラウド基盤上での実現について調査している.本稿では,仮想マシンを遠隔利用する場合の問題点と,実用する場合の通信帯域および RDP 性能について調査した結果を報告する. .
49. 伊東 栄典, Brendan Flanagan, YIN CHENGJIU, 中藤 哲也, 廣川 佐千男, クラウド型教育環境での検索エンジン構築演習, 電気学会 情報システム研究会, IS-13, 1-5, 2013.05, 近年,大学における情報機器を援用する教育活動や,情報科学技術の教育で,仮想化システムやクラウドシステムの利用が進んでいる。筆者らが所属する九州大学情報基盤研究開発セ ンターでは,2011年度末にCloudStackとVCLによるクラウドシステムを導入した。本稿では,導入したVCLシステムの構成を説明し,かつ検索エンジン構築演習に用いた事例を述べる。またVCLシステムを使った演習で得た知見を述べ,かつ問題点についての改善策を考察する。.
50. 伊東 栄典, 堀 良彰, 笠原 義晃, 井上 弘士, 情報系大学院でのVCL活用, 情報処理学会 情処研報, 2013-CLE-9, 9, 1-6, 2013.02, [URL], 近年,大学における情報機器を援用する教育活動や,情報科学技術の教育で,仮想化システムやクラウドシステムの利用が進んでいる.筆者らが所属する九州大学では, 2010年度末に九州大学システム情報科学研究院が同院内の教育用にVCLをベースとするシステムを導入した.また,2011年度末,九州大学情報基盤研究開発センターでは,主に大学院での教育に用いるための情報システムとしてVCLシステムを導入した.本稿では,導入したVCLシステムの構成と講義での利用事例を述べる.またVCLシステムの運用で得た様々な問題点について考察する.
.
51. Eisuke Ito, Kazunori Shimizu, Sachio Hirokawa, Predicting Future Ranking of Online Novels based on Collective Intelligence, Proc. of ICDIPC2013 (The Third Int'l Conf on Digital Info. Processing and Communications), SDIWC, 261-272, 2013.02, [URL], A large number of novels are being upload- ed as online novels. The present paper pro- poses 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..
52. 伊東 栄典, 清水 一憲, 廣川 佐千男, 集合知に基づくオンライン小説のランキング手法, 電子情報通信学会 信学技法, 112, 346, 107-112, 2012.12, [URL], オンライン小説サービスでは膨大な小説が投稿されている。本稿では,読者のお気入り登録(ブックマー ク)を用いて,新規投稿小説のうち,将来ランキング上位になる小説を早期に発見する手法を提案する。提案手法の評価実験を小説のジャンル毎に行なった。その結果、幾つかのジャンルについて,5月の時点でのブックマークデータから,7 月の人気上位の小説を推定することができた。.
53. Eisuke Ito, Kazunori Shimizu, Frequency and link analysis of online novels toward social contents ranking, The 2nd International Conference on Social Computing and its Applications (SCA2012), IEEE, 10.1109/CGC.2012.18, 531-536, 2012.11, [URL], User generated contents service is a social service because users not only enjoy contents but also provide feedback on content by commenting, tagging and bookmarking. The authors are interested in social service ranking and categorization methods for online novels. Many readers comment and bookmark there favorite novels in online novel services. Comments and bookmarking are facilitated by readers, and it is possible to use the data as a resource for social ranking and recommendation. In this paper, we focused on an online novel service, and analyzed the frequency of keywords, number of authors, and links from readers to novels. Although the bipartite graph between readers and novels fulfills growth and preferential attachment conditions of scale- free network, distribution of links does not follow power law. The basic idea of social ranking using extracted bookmarked novels and favorite authors is also explored..
54. Naomi Fujimura, Tadatsugu Togawa, Yoshiaki Kasahara, Eisuke Ito, Introduction and Experience with the Primary Mail Service based on their Names for Students, ACM SIGUCCS'12, 10.1145/2382456.2382460, 11-14, 2012.10, [URL], Kyushu University provided mail service based on student IDs such as “1AB10123X” to the university students for many years. Using this model, we had problems communicating with the stu-dents who graduated from the University and enrolled in the grad-uate school. The students received new mail addresses based on their new student IDs such as “2AB12789Y”. Faculty members were forced to change the student mail addresses in the mailing lists and in their mail client address book. Furthermore, students were forced to notify the e-mail address change to all of their existing contacts.
We introduced a new mail system to provide addresses based on the student name, as well as student ID in April 2011. The new naming convention uses the following format: “last-name.firstname.999” where “999” is a random number of 3 digits. Students can select some combination patterns of their first and last names for Japanese. We also consider the middle names for foreign students.
In the system implementation, we did not have the formal infor-mation of alphabetical names for students. We generated alphabet-ical names from Japanese Katakana names. It is not easy for us to get the appropriate name in this manner. We implemented a con-firmation stage of the alphabetical name at first use, and then stu-dents can select their mail addresses for their convenience. We paid much attention to the user interface in the system..
55. 笠原 義晃, 伊東 栄典, 堀 良彰, 藤村 直美, Gmailが大学メールサーバへ与える負荷状況の分析, 情報処理学会 情処研報, 2012-EVA-39, No.5, 2012.09, [URL], 九州大学では,従来から大学ドメインのメールサーバを学内に構築し,構成員へメールサービスを提供してきた. 2012 年 1 月頃より,学内の情報サービスに対し利用者認証機能等を提供する全学認証サーバの負荷の高さが問題とな り,その原因の一つが学生向けメールサーバであることが明らかになった.詳細な分析の結果,Google 社の Gmail から本学のメールサーバへ持続的なアクセスがあり,中でも既に卒業などで消滅したアカウントへのアクセスが多数 あることが分かった.本稿では,本学の学生メールサーバのアクセスログ解析に基づいて,Gmail がメールサーバへ 与える負荷状況の分析とその理由について述べ,対応策について検討する..
56. Eisuke Ito, Kazunori Shimizu, Sachio Hirokawa, Development of Facet Analysis System for Diverse Online Novels, DLINE Journal of Data Processing, 2, 3, 113-119, 2012.09, [URL], In recent years, user generated content services have become popular. The authors are interested in user generated 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 in classifying seaching 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 preference..
57. Hiroaki Ninomiya, Eisuke Ito, Brendan Flanagan, Sachio Hirokawa, Near friends communication encouragement system using NFC and SNS, AAI2012 (International Conference on Advanced Applied Informatics), 10.1109/IIAI-AAI.2012.37, 145-148, 2012.09, [URL], The value of an open event is in active communication between participants. 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 a 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..
58. Hiroaki Ninomiya, Eisuke Ito, Brendan Flanagan, Sachio Hirokawa, Bridging SNS ID and User Using NFC and SNS, Proc. of IEEE ASID2012 (The sixth IEEE International Conference on Anti-counterfeiting, Security, and Identification), 10.1109/ICASID.2012.6325346, 311-315, 2012.08, [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 on-line 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..
59. Eisuke Ito, Sachio Hirokawa, Kazunori Shimizu, Introducing faceted views in diversity of online novels, Proc. of IEEE ICDIM 2012 (Seventh International Conference on Digital Information Management), 10.1109/ICDIM.2012.6360114, 145-148, 2012.08, [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.
60. 伊東栄典, 笠原義晃, 藤村直美, 全学認証サーバの負荷状況と負荷分散, 情報処理学会 情処研報, 2012 CSEC-57/IOT-17, 11, 51-56, 2012.05, [URL], 近年,大学では全学的な認証基盤構築が進んでいる.筆者らが所属する九州大学でも全学共通IDの発行および認証基盤を構築し,学内向け情報サービスの利用者認証を一元化している.認証基盤を利用するサービスの増加に伴い,LDAP認証サーバの負荷が上昇している.中でも電子メール利用時の認証と,無線LAN接続時の認証が認証サーバの負荷を増大させている.本論文では,九州大学認証基盤における負荷状況の解析結果を述べる.また,負荷分散の方法を提案し,実際に負荷分散を適用した効果について,短期間ではあるものの,その結果を示す..
61. 村上直至, 伊東栄典, 動画コンテンツの視聴者コメントに基づくランキングとその評価, DEIM2012 (第4回データ工学と情報マネジメントに関するフォーラム), F8-3, 2012.03.
62. 阿部葉月, 伊東栄典, 橋本司, 高品質コンテンツ検索のための集合知の分散処理, DEIM2012 (第4回データ工学と情報マネジメントに関するフォーラム), E6-3, 2012.03.
63. Naomichi Murakami, Eisuke Ito, Emotional video ranking based on user comments, Proc. of iiWAS2011 (The 13th International Conference on Information Integration and Web-based Applications & Services), ACM, 499-502, 2011.12.
64. 伊東栄典, 上田将嗣, 九州大学キャンパスクラウドシステムの導入, 第33回 全国共同利用情報基盤センター 研究開発連合発表講演会, 109-114, ISSN 0910-8769, 2011.11, 高速ネットワークの普及,計算機性能の向上,仮想化技術およびWeb技術の進展から,クラウド技術が進展している。九州大学情報基盤研究開発センター(以下,センターと略記)では,主に大学院での情報環境を用いた教育研究に用いるためのプライベートクラウドシステムを導入し,各大学へ提供することを計画しており,その手始めとして,情報系の大学院への提供を想定したクラウドシステムを導入中である。本稿では導入中の九州大学キャンパスクラウドシステムについて報告する。.
65. Chengjiu Yin, Eisuke Ito, Tetsuya Nakatoh, Sachio Hirokawa, Classification network of tourism information -- A smart phone-based system for supporting "Petit Trips" --, Proc. of iCAST (3rd International Conference on Awareness Science and Technology), IEEE, 10.1109/ICAwST.2011.6163134, 170-173, 2011.09, [URL].
66. Kensuke Baba, Nami Hoshiko, Eriko Kudo, Naomi Yoshimatsu, Eisuke Ito, Semi-automated Paper-registration System for Institutional Repository, Proc. of iCAST (3rd International Conference on Awareness Science and Technology), IEEE, 10.1109/ICAwST.2011.6163111, 418-421, 2011.09, [URL].
67. Kensuke Baba, Maso Mori, Eisuke Ito, Identification of Scholarly Papers and Authors, Proc. of NDT2011 (Networked Digital Technologies), Springer LNCS CCIS136, 195-202, Springer LNCS CCIS136, 2011.07.
68. Kensuke Baba, Masao Mori, Eisuke Ito, Sachio Hirokawa, A Feedback System on Institutional Repository, Proc. of INTENSIVE 2011, IARIA, 37-42, 2011.05.
69. Kensuke Baba, Eisuke Ito, Sachio Hirokawa, Co-occurrence Analysis of Access Log of Institutional Repository, JCAICT2011 (Japan-Cambodia Joint Symposium on Information Systems and Communication Technology), 25-29, 2011.01.
70. 村上直至, 伊東栄典, 動画投稿サイトで付与された動画タグの階層化, 情報処理学会研究報告, 2010-MPS-81, 17, 2010.12.
71. 伊東栄典, 笠原義晃, 藤村直美, 九州大学全学基本メールの機能改善と有料サービスクラスの開始, 情報教育研究集会2011, B2-4, 2010.12.
72. 伊東栄典, 馬場謙介, 森雅生, 研究者DBと機関リポジトリとの連携機能開発, 情報教育研究集会2010, B3-6, 2010.12.
73. Kensuke Baba, Masao Mori, and Eisuke Ito, A Synergistic System of Institutional Repository and Researcher Database, Proc. of SERVICE COMPUTATION 2010, IARIA, 184-188, 2010.11.
74. 北祐一郎, 伊東栄典, 平野広幸, 吉松勝, 戸川忠嗣, マトリックス認証装置を使ったアクセス事例, 第32回 全国共同利用情報基盤センター 研究開発連合発表講演会, 81-87, ISSN 0910-8769, 2010.11.
75. 村上直至, 伊東栄典, 共感度検索のための動画アノテーション分析, FIT2010(第9回情報科学技術フォーラム), 第2分冊, 343-348, 2010.08.
76. 伊東栄典,片岡真,牧瀬ゆかり, Shibboleth認証基盤構築と学術認証フェデレーションへの参加 -今後のeリソースサービス基盤にむけて-, 九州大学附属図書館年報2009/2010, 11-15, ISSN: 1881-3542, 2010.08.
77. 伊東栄典, 馬場謙介, 吉松直美, 星子奈美, 機関リポジトリQIRの利用動向調査 -機関リポジトリの利用促進のために-, 九州大学附属図書館年報2009/2010, 16-18, ISSN: 1881-3542, 2010.08.
78. 阿部英司, 伊東栄典, 笠原義晃, 認証フェデレーションにおけるIdPの属性制御, 電子情報通信学会2010年総合大会講演論文集, B, S133-134, 2010.03.
79. 村上直至, 伊東栄典, 視聴者コメントに基づく動画検索手法, 電子情報通信学会2010年総合大会, D, 24, 2010.03.
80. 馬場謙介, 伊東栄典, 吉松直美, 星子奈美, 機関リポジトリの有効性分析, DEIM2010 (第2回データ工学と情報マネジメントに関するフォーラム), F7-3, 2010.03.
81. Kensuke Baba, Eisuke Ito, Naomi Yoshimatsu, Nami Hoshiko, Kazuaki Murakam, A Model of Publication of Scholarly Papers on Institutional Repositories , DRFIC2009, 2009.12.
82. 伊東栄典, 笠原義晃, 藤村直美 , 九州大学における職員向け電子メールサービスの現状 , 平成21年度情報教育研究集会 , D3-4, 2009.11.
83. Masanori NAKAKUNI, Hiroshi DOZONO, Eisulke ITO, Yoshiaki KASAHARA and Hideaki NAKAKUNI, A Method of Personal Authentication by Shape Recognition of the Lips and Front Teeth, Proc. of MACMESE'09 (MATHEMATICAL and COMPUTATIONAL METHODS), 203-207, 2009.11.
84. 戸川忠嗣, 宮口忠幸, 酒井健禎, 亀岡兼一, 入江啓一, 伊東栄典 , 九州大学における全学基本メールサービスの導入 , 第31回 全国共同利用情報基盤センター 研究開発連合発表講演会, 87-94, ISSN 0910-8769, 2009.11.
85. 阿部英司, 伊東栄典, 笠原義晃 , 要認証サイトのマッシュアップについての考察, 第62回電気関係学会九州支部連合大会 , 08-2P-07, 2009.09.
86. 園田亮, 伊東栄典 , 専門用語の関係に着目した医療文献検索手法に関する考察, 情報処理学会火の国情報シンポジウム2009, C-4-3, 2009.03.
87. 小野 昂, 阿部 英司, 中國 真教, 笠原 義晃, 伊東 栄典 , 分散SSO機構を用いたコミュニティ認可に関する考察 , 情報処理学会火の国情報シンポジウム2009 , B-6-2, 2009.03.
88. 場田 隼也,伊東 栄典, 中國 真教 , 大学における私的電子公証サービス, 電子情報通信学会2009総合大会 , BS-11-4, 2009.03.
89. 菅尾貴彦,戸川忠嗣,太田美和,橋倉聡,平野広幸,伊東栄典,市川広大,先立英喜 , 全学共通認証基盤サービスの手続きの電子化について , 第30回 全国共同利用情報基盤センター 研究開発連合発表講演会 , 30, 77-86, ISSN 0910-8769, 2008.11.
90. Masanori Nakakuni, Eisuke Ito, Yoshiaki Kasahara, Sozo Inoue, Hiroshi Dozono, Construction and Use Examples of Private Electronic Notary Service in Educational Institutions, WSEAS TRANSACTIONS on ADVANCES in ENGINEERING EDUCATION, 5, 10, ISSN: 1790-1979, 2008.10, [URL].
91. Masanori Nakakuni, Eisuke Ito, Yoshiaki Kasahara and Hiroshi Dozono , Private Electronic Notary Service in Universities and Its Utilization in Education , EDUTE'08 (The 4th WSEAS/IASME International Conference on EDUCATIONAL TECHNOLOGIES) , 170-175, 2008.08.
92. 阿部 英司, 伊東 栄典, 笠原義晃, 中國真教, 認証つきサービスにおける組織間連携のためのPKIとOpenIDの融合, 情報処理学会第二回インターネットと運用技術研究会 (情処研報 IOT-2-2008) , 17-22, 2008.07.
93. Masanori Nakakuni, Hiroshi Dozono, and Eisuke Ito , A Method of Automatic User Authentication by Fulltime Monitoring of Keystroke Timings, SAM'08 (The 2008 International Conference on Security and Management) , 58-63, 2008.07.
94. 島松 千春, 御手洗 秀一, 伊東 栄典, 廣川 佐千男, クチコミ情報からの関連商品マイニング, 電子情報通信学会第19回データ工学ワークショップ (DEWS 2008), B8-3, 2008.03.
95. 園田 亮, 伊東 栄典, 高崎 哲也, 川原 弘三, 医療文献からの類似項目発見についての考察, 電子情報通信学会第19回データ工学ワークショップ (DEWS 2008), C5-2, 2008.03.
96. 阿部 英司, 伊東栄典, PKI手法を用いた組織間での利用者認証連携ついての検討, 電子情報通信学会第19回データ工学ワークショップ (DEWS 2008), C1-6, 2008.03.
97. 百田信, 伊東 栄典, ソーシャルブックマークに基づく情報発見, 電子情報通信学会第19回データ工学ワークショップ (DEWS 2008), I1-15, 2008.03.
98. Eisuke Ito, Yoshiaki Kasahara, Megumi Nogita, Takahiko Suzuki, Institutional authentication platform for trustful inter/intra-institutional ubiquitous services, Proc. of the 2nd International Conference of Ubiquitous Information Technology, pp.103-108, 2007.12.
99. 伊東 栄典, のぎ田 めぐみ, 笠原 義晃, 鈴木 孝彦, 認証連携による無線LANローミング環境 -九州大学におけるUPKI・eduroamの連携-, 情報処理学会 研究会報告 2007-DPS-132/2007-GN-65/2007-EIP-37, 2007-DPS-132/2007-GN-65/2007-EIP-37, pp.141-146, 2007.09.
100. 園田亮, 伊東栄典, 池田大輔, 竇ギョクホウ, 笠原義晃, 大量の音楽プレイリストに基づく楽曲推薦システムの試作, 情報処理学会 第69回全国大会, 第二分冊, 189-190, 3N-5, 2007.03.
101. 百田信, 甲斐啓文, 伊東栄典, 分散環境におけるセキュアな協調作業のためのグループ管理システムの試作, 情報処理学会 第69回全国大会, 第三分冊, 99-100, 2U-1, 2007.03.
102. 冨山北斗, 伊東栄典, 廣川佐千男, 自己学習型トピッククローラーの性能評価に関する考察, 情報処理学会 第69回全国大会, 2007.03.
103. 島松千春, 杉本典子, 伊東栄典, 廣川佐千男, 田畑義之, 導出手続きによるXMLの自動変換, 情報処理学会 第69回全国大会, 第一分冊, 491-492, 5S-2, 2007.03.
104. 渦尾秀勝, 森雅生, 伊東栄典, 廣川佐千男, 双クラスタリングと概念束を用いた大規模文書検索法の提案, 情報処理学会 第69回全国大会, 2007.03.
105. 池田嗣穂, 亀岡謙一, 伊東栄典, RSSを利用した九大ポータルサイトの構築, 全国共同利用情報基盤センター・連合発表会, 2006.12.
106. 伊東 栄典, 利用者認証に用いる識別子の決定方法に関する考察, 電子情報通信学会 信学技報 ISEC2006-112, 67-72, 2006.12.
107. 冨山北斗, 下司義寛, 伊東 栄典, 廣川 佐千男, 自己学習型トピッククローラーの構築と評価, 電子情報通信学会 第17回データ工学ワークショップ(DEWS 2006), 3B-i11, 2006.03.
108. 伊東栄典, 島松千春, 廣川佐千男, 篠原正典, Webシラバス統合による教育情報ライブラリ構築, デジタル図書館, ISSN:1340-7287, pp.3--13, 2006.03.
109. 原由加里, 伊東栄典, 廣川佐千男, 大多和直樹, Web巻き − Web時代の絵巻物, 電子情報通信学会 IEICE SIG Notes WI2-2006-01, 1-5, 2006.01.
110. Yufeng Dou,, Eisuke Ito, Sachio Hirokawa, Daisuke Ikeda, An Approach to Analyzing Correlation between Songs/Artists Using iTMS Playlists, International Conference on Intelligent Agents, Web Technology andInternet Commerce (IAWTIC'2005), 951-956, 2005.12.
111. 廣川佐千男, 下司義寛, 池田大輔, 伊東栄典, Yufeng Dou, プレーリストからの曲目やアーティストの相互関係抽出, 情報処理学会FI研究会, 情処研究会報告Vol.2005, 108, 2005-FI-081, 31-36, 2005.11.
112. 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.
113. 渦尾秀勝, 甲斐啓文, 伊東栄典, 廣川佐千男, 学内ポータルサイト構築のための分散Web情報の一元化, 情報処理学会 マルチメディア,分散,協調とモバイルシンポジウム (DICOMO 2005), 465-468, 2005.07.
114. 甲斐啓文, 伊東栄典, 大庭淳一, 青柳睦, MO-SRVにおけるGrid計算サービスのための分散認証機構に関する考察, 情報処理学会第67回全国大会, 講演論文集(3), 719-720, 2005.03.
115. 伊東栄典, 竇鈺峰, 廣川佐千男, Webシラバス群のデータ形式統合に関する考察, マルチメディア,分散,協調とモバイルシンポジウム (DICOMO 2004), 345-348, 2004.07.
116. Sachio Hirokawa, Eisuke Ito, Tetsuhiro Miyahara, Semi-Automatic Construction of Metadata from A Series of Web Documents, Lecture Notes in Artificial Intelligence (LNAI 2903), 2903, 942-953, Springer LNAI 2903, pp.942-953, (ISSN 0302-9743, ISBN 3-540-20646-9), 2003.12.
117. 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, 電子情報通信学会和文論文誌D-I, Vol.J86, No.8, pp.566-574, 2003.08.
118. 伊東栄典, 松永吉広, 山田信太郎, 廣川佐千男, WebシラバスからのDB構成, 人工知能学会, 1D4-08, 2003.06.

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。