Tokushima University / Educator and Researcher Directory --- Tanioka, Hiroki

Personal Web Page

https://researchmap.jp/taniokah (researchmap)

https://orcid.org/0000-0003-3404-0855 (ORCID)

Field of Study

○	Information Security
○	Sports Science
○	Medical Informatics
○	Information Retrieval
○	Information Engineering

Subject of Study

○	Information Security Measures by Machine Learning (information security, machine learning, data mining)
○	Data Analysis using Statistics and Machine Learning (sports, machine learning, data mining)
○	Text Mining for Medical Data (information retrieval, text mining, medical infomatics)
○	Information Retrieval and Interactive Systems (information retrieval, information sharing, dialogue understanding)
○	Education on Information Science and Programming (programming, artificial intelligence, information processing)

Book / Paper

Book:

1.	Motohiro Ishida, Shinki Oyabu, Tetsushi Ueta, Hidekazu Kakei, Kazuhide Kanenishi, Hiroki Tanioka, Shin-ichi Nakayama and Akihiro Haga : 情報科学入門統計・データサイエンス・AI, 技術評論社, Mar. 2021.
2.	Hiroki Tanioka and XIN KANG : いちばんやさしいディープラーニング入門教室, Sotechsha Co., Ltd., Tokyo, Feb. 2018. (Keyword) machine learning / Deep Learning / Keras / Python
3.	Takebayashi Akira, Hiroki Tanioka and Kusano Shinichi : 12歳からはじめるJavaScriptとウェブアプリ, Rutles Inc., Tokyo, Nov. 2017.

Academic Paper (Judged Full Paper):

1.	Hiroki Tanioka, Mitsuhiro Sato and Ryuji Tsuge : Correlation Analysis between Impressions of Soccer School and Moving Speed in Elementary School Children, Information Engineering Express, Vol.9, No.2, 1-11, 2023. (Summary) In sports, the importance of enjoying exercise and interaction with others has been suggested, and it is an important question as to what kind of activities lead to enjoyment. The results of the questionnaire survey of soccer school students indicated that game-style practice is the most enjoyable, and that enjoyment was related to lots of running and lots of involvement with the ball. However, it was not clear how much they actually ran. Therefore, in this study, in order to analyze the relationship between the results of the questionnaire survey and actual movement, we analyzed the running style in game-style practice using GPS (Global Positioning System) and examined the correlation with the questionnaire results. The results showed that feeling that they ran well did not correlate well with actual distance or speed. On the other hand, moving at low speed are related to having fun and being able to shoot and pass the ball a lot. This result contradicts the survey results, which showed that students who responded that they enjoyed also responded that they ran a lot, making it clear that the results of the survey cannot be taken for granted. (Keyword) soccer / GPS / speed / correlation analysis (Tokushima University Institutional Repository) ● Metadata: 119013 (Link to Publication Site) ● Publication site (DOI): 10.52731/iee.v9.i2.768 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.52731/iee.v9.i2.768 (Tokushima University Institutional Repository: 119013, DOI: 10.52731/iee.v9.i2.768)
2.	Hiroki Tanioka, Mitsuhiro Sato and Ryuji Tsuge : Questionnaire Analysis and Causal Search for Enjoyment in Soccer Schools, IPSJ Transactions on digital practices, Vol.4, No.3, 74-87, 2023. (Summary) The importance of enjoying physical activity and interacting with others in sports has been sug- gested. This is an important theme for soccer schools, whose mission is to promote sports in the community. In our previous study, we conducted a questionnaire survey of students attending a soccer school and found that game-style practice leads to enjoyment. The results of the correlation analysis among the questionnaire items suggested that the enjoyment of soccer school is related to running well and being involved with the ball. However, it was not clear from the correlations whether the enjoyment came from running a lot and being involved with the ball more, or from running a lot and being involved with the ball more because it was enjoyable. Therefore, in this paper, we consider the questionnaire items collected using a 5-point Likert scale, such as enjoyed, ran a lot, and involved a ball a lot by shooting, passing, dribbling, etc., as continuous variables with a non-Gaussian distribution. To obtain causal relationships among the survey items, we attempted to conduct a causal search using DirectLiNGAM. The results of the causal search were consistent with the results of the correlation analysis, indicating that the respondents enjoyed soccer school, which may have led them to run a lot and get involved with the ball more. This practice clarified the causal relationship between the results of the questionnaire, which had not been clarified before, and provided useful information for the management of the school. (Keyword) correlation / causal exploration / LiNGAM / soccer (Tokushima University Institutional Repository) ● Metadata: 118184 (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1050296808060877952 (Tokushima University Institutional Repository: 118184, CiNii: 1050296808060877952)
3.	Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Tomohito Wada and Naka Gotoda : Double-sided design for reinforcing private training of basketball with an advanced imagery opponent, International Journal of Learning Technologies and Learning Environments, Vol.6, No.1, 18pgs, 2023. (Keyword) Mental imagery training / basketball / one-to-one / visualized opponent (Link to Publication Site) ● Publication site (DOI): 10.52731/ijltle.v6.i1.718 (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1390295956282570112 ● Search Scopus @ Elsevier (DOI): 10.52731/ijltle.v6.i1.718 (DOI: 10.52731/ijltle.v6.i1.718, CiNii: 1390295956282570112)
4.	Rina Yano, Hiroki Tanioka, Kenji Matsuura, Masahiko Sano and Tetsushi Ueta : Quantitative Measurement and Analysis to Computational Thinking for Elementary Schools in Japan, Information Engineering Express, IIAI International Journal Series, Vol.8, No.1, IEE658, 1-17, 2022. (Summary) In Japan, programming education has been made compulsory in elementary schools since 2020. The Programming Education Guide (GPE) explains the purpose of programming education and the abilities that can be fostered through programming education. In addition, the Portal Site for Programming Education Focusing on Elementary Schools introduces various examples of programming education. However, there is little information measuring whether programming classes are effective in improving OTWP (Objective Thinking as a Way of Programming) abilities based on CT (Computational Thinking), except for reports of improvement after simple statistical analysis. Therefore, we prepared 30 CT questions, 12 basic and 18 applied, for the CT test considering four key techniques, decomposition, pattern recognition, abstraction, and algorithms, of which 14 questions were pre-test and seven questions were assessment test. In the experiment, 18 elementary school students from grades 1st to 6th were given a short workshop only once, and the analysis of the effect was done statistically, considering their habituation to the problems. The results of the experiment showed that there was no effect of the one-time workshop, unlike other reports of improvement that used simple statistical methods. It became clear that the CT ability was not improved by the short education. On the other hand, a new finding is that females may be inferior to males in three techniques: decomposition, algorithm, and abstraction. (Keyword) Programming education / Computational thinking / Learning analytics (Tokushima University Institutional Repository) ● Metadata: 116887 (Link to Publication Site) ● Publication site (DOI): 10.52731/iee.v8.i1.658 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.52731/iee.v8.i1.658 (Tokushima University Institutional Repository: 116887, DOI: 10.52731/iee.v8.i1.658)
5.	Hiroki Tanioka : AI Boom Analysis Using Time Series Data in Word Embeddings, Society for Research on Work Competency, No.8, 2-15, 2020. (Summary) Replacing many of "work" of humanity with Artificial Intelligence (AI) is said to bring a big change in the industrial structure. However, this change will create a new "work". It will be evident from the change of "work" in the past industrial revolution. This paper describes how artificial intelligence, one of the key factors of the Fourth Industrial Revolution, has become recognized and noticed in Japanese society. We clarify by extracting time-series data using word embeddings from newspaper articles, etc. (Keyword) artificial intelligence / Word Embeddings / Time Series Data / text mining (Tokushima University Institutional Repository) ● Metadata: 114935 (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1050865122805897216 (Tokushima University Institutional Repository: 114935, CiNii: 1050865122805897216)
6.	Shu Yano, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Naoya Kohda, Naka Gotoda and Tomohito Wada : A learning environment of a systematic and tactical style in field sports, --- Basic tactic of an offense team in basketball ---, Transactions of Information Processing Society of Japan, Vol.61, No.3, 657-666, 2020. (Keyword) Tactics comprehension / Team match field sports / computer simulation / basketball (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1050283687644442112 (CiNii: 1050283687644442112)
7.	Tomohiro Takeda, Kenji Ohira, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : A Method to Detect Rogue Access Points in a Campus without Decrypting WLAN Frames, The Transactions of the Institute of Electronics, Information and Communication Engineers B, Vol.J101-B, No.2, 90-99, 2018. (Summary) Security problems of rogue APs exist because there are many wireless access points (APs) in a campus. In many cases, there are not only university-wide network administrators but also subnet administrators and the former ones usually do not know details of each subnet. We propose a method for the university-wide network administrators to detect rogue APs without decrypting WLAN frames. HTTP packets of captive portal detection are used in this method. As a result of our evaluation, we confirmed that it is possible to associate an IP address as seen from wired upstream network and a frame from WLAN. (Keyword) campus network / rogue access point / captive portal / wireless LAN (Tokushima University Institutional Repository) ● Metadata: 113976 (Link to Publication Site) ● Publication site (DOI): 10.14923/transcomj.2017GTP0013 (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1050865122806271744 ● Search Scopus @ Elsevier (DOI): 10.14923/transcomj.2017GTP0013 (Tokushima University Institutional Repository: 113976, DOI: 10.14923/transcomj.2017GTP0013, CiNii: 1050865122806271744)
8.	Hiroki Tanioka, Takafumi Bando and Kenji Matsuura : Considering adopting a full-text search on the Center for Administration of Information Technology, Tokushima University, Journal for Academic Computing and Networking, Vol.20, 30-39, 2016. (Keyword) 情報系システム / ISMS / 情報検索 / Elasticsearch (Link to Publication Site) ● Publication site (DOI): 10.24669/jacn.20.1_30 (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1390845712994884736 ● Search Scopus @ Elsevier (DOI): 10.24669/jacn.20.1_30 (DOI: 10.24669/jacn.20.1_30, CiNii: 1390845712994884736)
9.	Hiroki Tanioka : Pseudo Relevance Feedback Using Fast XML Retrieval, Advances in Focused Retrieval, 7th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2008, Dagstuhl Castle, Germany, December 15-18, 2008. Revised and Selected Papers, Vol.5631 LNCS, 218-223, 2009. (Summary) This paper reports the result of experimentation of our approach using the vector space model for retrieving large-scale XML data. The purposes of the experiments are to improve retrieval precision on the INitiative for the Evaluation of XML Retrieval (INEX) 2008 Adhoc Track, and to compare the retrieval time of our system to other systems on the INEX 2008 Efficiency Track. For the INEX 2007 Adhoc Track, we developed a system using a relative inverted-path (RIP) list and a Bottom-UP approach. The system achieved reasonable retrieval time for XML data. However the system has a room for improvement in terms of retrieval precision. So for INEX 2008, the system uses CAS titles and Pseudo Relevance Feedback (PRF) to improve retrieval precision. (Link to Publication Site) ● Publication site (DOI): 10.1007/978-3-642-03761-0_22 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-70350486211 (DOI: 10.1007/978-3-642-03761-0_22, Elsevier: Scopus)
10.	Hiroki Tanioka, Nakagawa Takashi and Maruyama Minoru : Improvement of Feature Extraction for Bayesian Spam Filtering, 情報処理学会論文誌数理モデル化と応用(TOM), Vol.1, No.1, 175-184, 2008. (Summary) We propose an improved baysian filter for spam mail detection. Bayesian filter was used on existing learning spam filters which achieved some positive results. Although we expect them to improve the true negative rate while keeping the false positive rate low. Therefore, it was based on a thorough review of distribution for each word and mail that our means of spam mail detection showed an impressively higher accuracy than ever. (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1050845762821854080 (CiNii: 1050845762821854080)
11.	Hiroki Tanioka : A Fast Retrieval Algorithm for Large-Scale XML Data, Focused Access to XML Documents, 6th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2007, Dagstuhl Castle, Germany, December 17-19, 2007. Selected Papers, Vol.4862 LNCS, 129-137, 2008. (Summary) This paper proposes a novel approach for retrieving large-scale XML data using the vector space model. The vector space model is commonly used in the information retrieval community. Last year, for the Evaluation of XML Retrieval (INEX) 2006 Adhoc Track, we developed a system using fragment elements. The system made it possible to search over XML elements for queries with varying constraints on XML elements to be included in the search, without the need for reindexing the collection, supporting more flexible queries. However the system took significant time to unitize the fragment elements. To solve the problem, our new system is composed of an inverted-file list and a relative inverted-path list on the INEX 2007 Adhoc Track corpus. (Link to Publication Site) ● Publication site (DOI): 10.1007/978-3-540-85902-4_12 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-51849148033 (DOI: 10.1007/978-3-540-85902-4_12, Elsevier: Scopus)
12.	Hiroki Tanioka : A Method of Preferential Unification of Plural Retrieved Elements for XML Retrieval Task, Comparative Evaluation of XML Information Retrieval Systems, 5th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2006, Dagstuhl Castle, Germany, December 17-20, 2006, Revised and Selected Papers, Vol.4518 LNCS, 45-56, 2007. (Summary) We developed a passage retrieval system for XML documents using the vector space model. To be more flexible for the query, we also developed a method of unification of multiple retrieved elements and a fragment indexing system. Our system is composed of an inverted file and an XML Path Language (XPath) path list. The validity of the method was tested as part of the ad hoc track in the Initiative for the Evaluation of XML Retrieval (INEX) 2006. (Link to Publication Site) ● Publication site (DOI): 10.1007/978-3-540-73888-6_5 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-38049102973 (DOI: 10.1007/978-3-540-73888-6_5, Elsevier: Scopus)

Academic Paper (Unrefereed Paper):

1.	Hiroki Tanioka, Kinya Tokunaga, Mikako Tokunaga, Yuki Ito, Haruka Kaneko, Reiko Miyamoto, Yu Kobayakawa, Nazuna Kobayashi and Kairi Ohara : Trend analysis of consumer issues using PIO-NET - Comparing topics in time series data using text mining, Research Discussion Paper, International Consumer Policy Research Center(ICPRC), Consumer Affairs Agency, 2023. (Summary) 本稿は，2021 年 1 月から 2021 年 12 月の 1 年間に受け付け，全国消費生活情報ネットワークシステム(PIO-NET)へ登録された相談内容のうち，新型コロナウイルス感染症関連の話題について抽出したものについて独自に分析し，消費者問題の傾向分析を行ったものである．PIO-NET の相談内容に含まれる話題の分析には，統計分析手法のトピックモデリングを用いる．話題の分類と特徴的なキーワードの抽出を行い，月ごとの新型コロナウイルス感染症関連の主な出来事について分析を行った．その結果，新型コロナウイルス感染症の流行が継続し，会社や家賃といった収入に関する相談や，ワクチン接種のような国の施策に関連する相談が目立った．同様の手法で 2020 年 1 月から 2020 年 12 月の 1 年間に受け付けた相談内容について分析を行った結果と比較した結果，新型コロナウイルス感染症の流行が始まったばかりの混乱期である 2020 年に多かったマスク等の商品の品不足関連の相談は 2021 年には目立たなくなり，送り付け商法等に関する相談や副業や会社，家賃といった経済的な理由に基づく相談が多くなったことが分かった．国の施策に関する相談についても，2020 年に多かった給付金やGo To トラベル等の相談に代わり，2021 年にはワクチン接種関連の相談が増加した．さらに，2020 年 1 月から 2021 年 12 月の 2 年間の相談内容について，トピックモデリングで得られたトラブルキーワードを用いて，連続する 2 ヶ月のトピックの類似度を計算することにより，トピックの変動を分析し，新型コロナウイルス感染症に関連する主な出来事との関連を考察した結果，緊急事態宣言の発出の前後では，大きくトピックの内容が変化している等，国の施策が消費者の相談内容に少なからず影響を与えていることが分かった． (Tokushima University Institutional Repository) ● Metadata: 118651 (Tokushima University Institutional Repository: 118651)
2.	Hiroki Tanioka, Kinya Tokunaga, Mikako Tokunaga, Mimako Chikamori, Tatsuya Mitani, Reiko Miyamoto, Ayumi Yamaguchi, Yu Kobayakawa and Anna Yasui : Trend analysis of consumer issues using PIO-NET, Research Discussion Paper, International Consumer Policy Research Center(ICPRC), Consumer Affairs Agency, 2022. (Summary) 本稿は，2020 年の 1 年間に全国消費生活情報ネットワークシステム(PIO-NET)へ登録された相談内容のうち，新型コロナウイルス関連の話題を抽出したものについて，傾向分析を行ったものである．PIO-NET の相談内容に含まれる話題の分析には，トピックモデリングと呼ばれる統計手法を用いて話題の分類と特徴的なキーワードの抽出を行う．また，各月ごとの新型コロナウイルス関連の主な状況について，消費生活相談の傾向と比較する．その結果，個別の話題の増減のタイミングで，消費生活相談も増減しており，大きな傾向の違いは見られなかったが，4 月下旬に関連する報道や施策が発表された特別定額給付金等について，消費生活相談としてはその 1 か月遅れでトピックとして現れるなど，政策の発表から相談として現れるまで，若干のタイムラグがあることが示唆される現象が見られた．また，トピック分析から得られたトラブルキーワードについて，その出現頻度や出現割合を分析した結果，そのトラブルキーワードが代表する相談が顕著に見て取れた．比較のために，同期間における日本経済新聞の朝刊で取り上げられた記事についても分析を行った．新聞記事データは，新型コロナウイルス関連のみを抽出したものではないため，新聞記事データを用いたトピック分析では，PIO-NET データのトピック分析結果とは全く異なる傾向であった．しかし，PIO-NET データの分析から得られたトラブルキーワードを用いて，新聞記事データにおける出現頻度や出現割合を分析することで，消費生活相談で話題となっているトピックの新聞報道における注目状況を概観することができた． (Tokushima University Institutional Repository) ● Metadata: 116909 (Tokushima University Institutional Repository: 116909)
3.	Hiroki Tanioka, Kenji Matsuura, Tetsushi Ueta and Fumiaki Kawano : Challenges and Improvements in a Workshop-Type Class in BYOD, Journal of University Education Research, No.16, 18-23, 2019. (Keyword) BYOD / workshop / group learning / information literacy (Tokushima University Institutional Repository) ● Metadata: 113256 (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1050865122806501888 (Tokushima University Institutional Repository: 113256, CiNii: 1050865122806501888)

Academic Letter:

1.	Hiroki Tanioka, Mitsuhiro Sato and Ryuji Tsuge : Analyze the enjoyment of soccer school in the elementary school age, IIAI Letters on Informatics and Interdisciplinary Research, Vol.1, 1-11, 2022. (Summary) The importance of enjoying physical exercise and interaction with others in sports has been suggested, however it has not been clarified what activities lead to enjoyment. Therefore, we conducted an online questionnaire survey of students attending soccer school after practice and conducted a correlation analysis of their impressions of attending school, whether they enjoy soccer school activities, and what activities lead to enjoyment, and found that the main factor contributing to their enjoyment of soccer school was game (match), also suggesting that enjoying school is related to running a lot and being involved with the ball a lot. On the other hand, students who could juggle well tended not to enjoy it. (Keyword) Correlation analysis / Enjoyment / Soccer (Tokushima University Institutional Repository) ● Metadata: 117448 (Link to Publication Site) ● Publication site (DOI): 10.52731/liir.v001.011 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.52731/liir.v001.011 (Tokushima University Institutional Repository: 117448, DOI: 10.52731/liir.v001.011)

Review, Commentary:

1.	Hiroki Tanioka : Data and Artificial Intelligence utilization required for sports analytics, The Journal of Information and Systems in Education, Vol.37, No.3, 192-197, Jul. 2020. (Summary) Tokyo 2020 Olympic Games has been postponed until 2021. Most of the 33 sports still planned for the Olympic Games in 2021 will use data. The sports data gathered using various method is analyzed by experts. The experts also called sports data analysts have been developed various systems and methods using the sports data and Artificial Intelligence to improve competition results. In this paper, data collections in some sports are explained, and the trends in research on the use of data analysis models in these fields, in particular, some studies using machine learning and our research are introduced. Additionally, future issues in this field are summarized. (Keyword) Sports Analytics / Data Science / artificial intelligence / machine learning (Tokushima University Institutional Repository) ● Metadata: 115142 (Link to Publication Site) ● Publication site (DOI): 10.14926/jsise.37.192 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.14926/jsise.37.192 (Tokushima University Institutional Repository: 115142, DOI: 10.14926/jsise.37.192)

Proceeding of International Conference:

1.	Hiroki Tanioka, Tetsushi Ueta and Masahiko Sano : Toward a Dialogue System Using a Large Language Model to Recognize User Emotions with a Camera, The 1st InterAI: Interactive AI for Human-Centered Robotics workshop in conjuction with IEEE Ro-MAN 2024, Pasadona, LA, USA, Aug. 2024. (Summary) The performance of ChatGPT© and other LLMs has improved tremendously, and in online environments, they are increasingly likely to be used in a wide variety of situations, such as ChatBot on web pages, call center operations using voice interaction, and dialogue functions using agents. In the offline environment, multimodal dialogue functions are also being realized, such as guidance by Artificial Intelligence agents (AI agents) using tablet terminals and dialogue systems in the form of LLMs mounted on robots. In this multimodal dialogue, mutual emotion recognition between the AI and the user will become important. So far, there have been methods for expressing emotions on the part of the AI agent or for recognizing them using textual or voice information of the users utterances, but methods for AI agents to recognize emotions from the users facial expressions have not been studied. In this study, we examined whether or not LLM-based AI agents can interact with users according to their emotional states by capturing the user in dialogue with a camera, recognizing emotions from facial expressions, and adding such emotion information to prompts. The results confirmed that AI agents can have conversations according to the emotional state for emotional states with relatively high scores, such as Happy and Angry. (Keyword) Multimodal Dialogue / Large Language Models / Facial Expression Recognition / Emotion Detection / ChatBot
2.	Kenji Matsuura, Hironori Takeuchi, Kazuki Urushihara, Hiroki Tanioka, Stephen Githinji Karungaru and Tomohito Wada : Multiplayer Tracking with Diagonal Video to Support Basketball Tactical Learning, 18th International Conference on Ubiquitous Information Management and Communication, 5pgs, Kuala Lumpur, Malaysia, Jan. 2024. (Summary) Player tracking has become a crucial technique for getting strategic advantage in competitive team sports. Movies captured by one or a few cameras are easy to use for this purpose at any level including amateur. This study describes a design issue that integrates up-to-date technologies of multi-object tracking for tactics detection from the actual game data to bring affordability, feasibility, and accessibility. (Keyword) multi-object-tracking / basketball / tactics / learning (Link to Publication Site) ● Publication site (DOI): 10.1109/IMCOM60618.2024.10418315 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85186143229 (DOI: 10.1109/IMCOM60618.2024.10418315, Elsevier: Scopus)
3.	Stephen Githinji Karungaru, Kenji Matsuura and Hiroki Tanioka : Basketball Players Identification and Tracking using a Single Fixed Camera, 17th International Conference on Signal Image Technology & Internet based Systems, 342-347, Bangkok, Nov. 2023. (Summary) The application of data science in sports has recentlygained significant importance. Consequently, tactics development,opponent team understanding, in-game decision-making, playervital statistics, etc. could all benefit from game data analysis.Given a video input, this study proposes employing consecutiveintersection over union (IOU) based tracking and K-Meansbased uniform colors detection methods to monitor players andidentify teams in basketball. A retrained YOLOv7 algorithmis used to recognize the humans (players, referees, etc.), ball,and field markers. The Kalman filter is employed to tackleocclusion occurrence during tracking. Our findings indicate thatthis technology may be practical and affordable for both amateurand professional teams. (Keyword) multi-object-tracking / basketball / tactics / machine learning (Link to Publication Site) ● Publication site (DOI): 10.1109/SITIS61268.2023.00062 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.1109/SITIS61268.2023.00062 (DOI: 10.1109/SITIS61268.2023.00062)
4.	Hiroki Tanioka, Tsuyoshi Miura, Kenji Matsuura and Stephen Githinji Karungaru : Development of semi-automatic image annotation using object recognition, Proceedings of IIAI-AAI2023, Koriyama, Jul. 2023. (Summary) The modern field of machine learning has made remarkable progress, and the demand for training data required by machine learning systems is constantly increasing. However, the annotation work required to create this training data requiresa great deal of effort. Therefore, we considered semi-automating the labeling process by taking advantage of false positives that occur when object detection is performed by YOLOv5, thereby eliminating the need for region selection and reducing the cost of annotation work. In a comparison experiment with labelImg, a common annotation tool, we found that semi-automating the annotation process took less time than manually annotating the objects. (Keyword) machine learning / image recognition / annotation / labeling (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI59060.2023.00109 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.1109/IIAI-AAI59060.2023.00109 (DOI: 10.1109/IIAI-AAI59060.2023.00109)
5.	Yuri Lucas Direbieski, Hiroki Tanioka, Kenji Matsuura, Hironori Takeuchi, Masahiko Sano and Tetsushi Ueta : Security Impact Analysis of Degree of Field Extension in Lattice Attacks on Ring-LWE Problem, COMPSAC 2023 Workshop: The 1st IEEE International Workshop on Data Science & Machine Learning for Cybersecurity, IoT & Digital Forensics (DSML), 1441-1446, Turin, Jun. 2023. (Summary) Modern information communications use cryptography to keep the contents of communications confidential. RSA (RivestShamirAdleman) cryptography and elliptic curve cryptography, which are public-key cryptosystems, are widely used cryptographic schemes. However, it is known that these cryptographic schemes can be deciphered in a very short time by Shors algorithm when a quantum computer is put into practical use. Therefore, several methods have been proposed for quantum computer-resistant cryptosystems that cannot be cracked even by a quantum computer. A simple implementation of LWE-based lattice cryptography based on the LWE (Learning With Errors) problem requires a key length of O(n2) to ensure the same level of security as existing public-key cryptography schemes such as RSA and elliptic curve cryptography. In this paper, we attacked the Ring-LWE (RLWE) scheme, which can be implemented with a short key length, with a modified LLL (Lenstra-Lenstra-Lova sz) basis reduction algorithm and investigated the trend in the degree of field extension required to generate a secure and small key. Results showed that the lattice-based cryptography may be strengthened by employing Cullen or Mersenne prime numbers as the degree of field extension. (Keyword) lattice-cipher / ring-LWE / LLL-basis-reduction (Link to Publication Site) ● Publication site (DOI): 10.1109/COMPSAC57700.2023.00221 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.1109/COMPSAC57700.2023.00221 (DOI: 10.1109/COMPSAC57700.2023.00221)
6.	Kenji Matsuura, Kohta Jobe, Hiroki Tanioka, Hironori Takeuchi and Tetsushi Ueta : Performed-Tactics Detection of a Basketball Match using Multi-Object Tracking Technology Applicable for a Movie, Elizabeth Langran (Ed.), Proceedings of Society for Information Technology & Teacher Education International Conference, 676-680, New Orleans, LA, US, Mar. 2023.
7.	Stephen Githinji Karungaru, Hiroki Tanioka and Kenji Matsuura : Soccer Players Real Location Determination Using Perspective Transformation, Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems (SCIS&ISIS2022), 1-4, Nov. 2022. (Keyword) pattern recognition / image processing / deep learning (Link to Publication Site) ● Publication site (DOI): 10.1109/SCISISIS55246.2022.10001939 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85146676613 (DOI: 10.1109/SCISISIS55246.2022.10001939, Elsevier: Scopus)
8.	Hiroki Tanioka, Mitsuhiro Sato and Ryuji Tsuge : Analyze the enjoyment of soccer school in the elementary school age, IIAI Letters on Informatics and Interdisciplinary Research (Proceedings of IIAI-AAI2022), Vol.1, 1-11, Online, Jul. 2022. (Summary) The importance of enjoying physical exercise and interaction with others in sports has been suggested, however it has not been clarified what activities lead to enjoyment. Therefore, we conducted an online questionnaire survey of students attending soccer school after practice and conducted a correlation analysis of their impressions of attending school, whether they enjoy soccer school activities, and what activities lead to enjoyment, and found that the main factor contributing to their enjoyment of soccer school was game (match), also suggesting that enjoying school is related to running a lot and being involved with the ball a lot. On the other hand, students who could juggle well tended not to enjoy it. (Keyword) Correlation analysis / Enjoyment / Soccer (Link to Publication Site) ● Publication site (DOI): 10.52731/liir.v001.011 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.52731/liir.v001.011 (DOI: 10.52731/liir.v001.011)
9.	Kenji Matsuura, Hiroki Tanioka, Naka Gotoda and Tomohito Wada : Self-studying environment with imagery rehearsal for a ball carrier in basketball, 2022 16th International Conference on Ubiquitous Information Management and Communication, 1-7, Online, Jan. 2022. (Summary) This study is focused on a basketball player whocarries the ball and the development of a learning supportenvironment for the players selected actions. Further, the basicactions for the selection in this study include passing, dribbling,and shooting as the targeting actions for the next time. Weapplied the Delaunay diagram, which is derived from a twodimensional overhead perspective of the court to automaticallyextract possible options for the next actions for learners toselect an action according to the scenario. The system appliedan estimation indicator with respect to shooting possibilities tothe extracted options. The indicator is determined based onthe positions of the ball carrier, other players, and the goalrepresented in the game data. This indexed value is called MyRecommendation, and it indicates the degree of potential success.We developed a system to show a learner the extracted actionoptions and the degree of recommendation. When a 2D animationis displayed in this system, the action options are indicated bycolored arrows starting from the ball carrier to the target option.Learners can understand typical behavioral patterns throughprototype use. (Link to Publication Site) ● Publication site (DOI): 10.1109/IMCOM53663.2022.9721766 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85127475362 (DOI: 10.1109/IMCOM53663.2022.9721766, Elsevier: Scopus)
10.	Kenji Matsuura, Hiroki Tanioka, Tomohito Wada and Naka Gotoda : A guide for inexperienced players of basketball to master basic field of vision, Proceedings of IIAI-AAI2021, 196-199, Online, Jul. 2021. (Summary) This paper describes a methodology of trainingconscious view for players of basketball. In order for thispurpose, we designed and implemented a learning support systemarranged from original 2D-simulator. The system provides bird's eye-view of the court where each defense in the simulator hasthe field guide of view angle. Through an experimental use, oursystem brought positive effect in learning basic view of playerstaking a role in defense. (Keyword) Skill Development / Field of View / Basketball (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI53430.2021.00035 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85133183028 (DOI: 10.1109/IIAI-AAI53430.2021.00035, Elsevier: Scopus)
11.	Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Tomohito Wada and Naka Gotoda : Design of a one-on-one training system for basketball players, Proceedings of IIAI-AAI2021, 135-140, Online, Jul. 2021. (Summary) This paper describes the design of a system to developthe feint skill of basketball players in one-on-one situations.When training alone for one-on-one situations, the player lacksan opponent; therefore such training lacks a realistic aspect.The proposed system uses a software to provide a single traineewith an appropriate opponent. To train an offence player, thesystem monitors and analyzes the body movements of an offensiveplayer, providing a counterpart movement with the silhouette ofa visualized defense, whose functions are based on the analysis ofthe monitored offense movement. To train a defense, the system ismodified to provide a realistic offence using a head mount displayworn by a defense trainee. Defense trainee receives feedback afterthe training. (Keyword) Skill Development / One-on-one / Basketball / Sensor (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI53430.2021.00024 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85133174599 (DOI: 10.1109/IIAI-AAI53430.2021.00024, Elsevier: Scopus)
12.	Hiroki Tanioka and Ryota Kohri : Improving the winning percentage of the Werewolf team through collusion strategies, Proceedings of IIAI-AAI2021, 944-945, Online, Jul. 2021. (Summary) In recent years, the number of systems using arti- ficial intelligence (AI) is increasing. Even in Chess, Shogi, and Go, which are perfect information games, AI has been used to beat masters. Furthermore, AI has been actively researched in the imperfect information games Poker and Mahjong, and has reached the point where it can be played on par with humans. On the other hand, the werewolf game focused on in our research is an incomplete information game, and it is also necessary to aim for victory as a team using asymmetric information. In this paper, we report that the winning percentage of werewolf team can be improved by taking a team strategy in a 15-player werewolf game. (Keyword) werewolf game / team strategy / artificial intelligence (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI53430.2021.00175 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85133181596 (DOI: 10.1109/IIAI-AAI53430.2021.00175, Elsevier: Scopus)
13.	Hiroki Tanioka and Rina Yano : Development and Evaluation of Quizzes Aimed at Quantifying Computational Thinking, Proceedings of IIAI-AAI2021, 188-191, Online, Jul. 2021. (Summary) Programming education will be compulsory in elementary schools in Japan from 2020, but it is still little information on how to measure the effects of lessons conducted for students. In this study, to measure objectively and quantitatively the educational effect of programming education, we developed 30 quizzes of measuring the ability of computational thinking. In this paper, focusing on the programming experience of college students and above, we confirmed the measurable ability using these quizzes. (Keyword) learning analytics / programming education / computational thinking (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI53430.2021.00033 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85133203214 (DOI: 10.1109/IIAI-AAI53430.2021.00033, Elsevier: Scopus)
14.	Rina Yano, Hiroki Tanioka, Kenji Matsuura, Masahiko Sano and Tetsushi Ueta : Quantitative Measurement and Analysis to Thinking as a Way of Programming for Elementary School in Japan, Proceedings of IIAI-AAI2020, 163-168, Kitakyushu, Sep. 2020. (Summary) Programming education will be compulsory in elementary schools in Japan from 2020, but it is still little information on how to measure the effects of lessons conducted for students. In this study, to measure objectively and quantitativelythe educational effect of programming education, we developed a test of measuring the ability of thinking as a way of programming. In this paper, we evaluate the measurability of the test using Fishers exact test for elementary school students answers, focusing on the grade, the programming experience, and the difference between the results before and after the workshop. (Keyword) Learning Analytics / Programming Education / Elementary School (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI50415.2020.00040 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85107198127 (DOI: 10.1109/IIAI-AAI50415.2020.00040, Elsevier: Scopus)
15.	Shu Yano, Kenji Matsuura, Hiroki Tanioka, Hiromu Naito, Naka Gotoda and Tomohito Wada : A Supporting System Design for Basketball Offense Tactics, Proceedings of IIAI-AAI2020, 213-216, Kitakyushu, Sep. 2020. (Summary) In terms of basketball, a player always decides an option of the next movement what leads to get score. In this research, we propose an environment focusing on a ball holder, who has to select the next action. As a support, our environment visualizes options for the ball holders next actions. Furthermore, the positional relationship among players is measured from the coordinate obtained from the game data, and the system caliculates and provides information about how easy to get score. We design such a system with these functions so that the user can learn how to select the next movement. (Keyword) support / basketball / tactics / learning (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI50415.2020.00049 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85107184803 (DOI: 10.1109/IIAI-AAI50415.2020.00049, Elsevier: Scopus)
16.	Stephen Githinji Karungaru, Kenji Matsuura, Hiroki Tanioka, Naoya Kohda, Naka Gotoda and Tomohito Wada : Towards Drone-Video Player Detection and Tracking for Soccer Strategy Analysis, Proceedings of the 6th IEEJ international workshop on Sensing, Actuation, Motion Control, and Optimization, Vol.TT-1, 1-6, Tokyo, Mar. 2020. (Keyword) Human Tracking / Drone / Tiny Objects / Particle Filter
17.	Shu Yano, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Naoya Kohda, Naka Gotoda and Tomohito Wada : Tactics-Trend Analysis for Increasing the Possibility of Shooting in a Basketball Match, Proceedings of IEEE IMCOM2020, 1-4, Taichung, Taiwan, Jan. 2020. (Link to Publication Site) ● Publication site (DOI): 10.1109/IMCOM48794.2020.9001784 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85081137407 (DOI: 10.1109/IMCOM48794.2020.9001784, Elsevier: Scopus)
18.	Hiroki Tanioka : A Fast Content-Based Image Retrieval Method Using Deep Visual Features, ICDAR-WML: The 2nd International Workshop on Machine Learning 2019, 20-23, Sydney, Sep. 2019. (Summary) Fast and scalable Content-Based Image Retrieval using visual features is required for document analysis, Medical image analysis, etc. in the present age. Convolutional Neural Network (CNN) activations as features achieved their outstanding performance in this area. Deep Convolutional representations using the softmax function in the output layer are also ones among visual features. However, almost all the image retrieval systems hold their index of visual features on main memory in order to high responsiveness, limiting their applicability for big data applications. In this paper, we propose a fast calculation method of cosine similarity with L2 norm indexed in advance on Elasticsearch. We evaluate our approach with ImageNet Dataset and VGG-16 pre-trained model. The evaluation results show the effectiveness and efficiency of our proposed method. (Keyword) Content-Based Image Retrieval / Deep Convolutional Representations / Bag of Visual Words / Elasticsearch (Link to Publication Site) ● Publication site (DOI): 10.1109/ICDARW.2019.40075 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.1109/ICDARW.2019.40075 (DOI: 10.1109/ICDARW.2019.40075)
19.	Kohei Kawanaka, Felicia Misamanana Rabarison, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : Improving Animal Recognition Accuracy using Image De-fencing, The 7th International Conference on Smart Computing and Artificial Intelligence (SCAI 2019), 534-537, Toyama, Jul. 2019. (Summary) When a computer recognizes an animal on an image, an accuracy of animal recognition decreases if a part of the animals body is covered by fences. On the other hand, when a human recognizes an object on the image, it is known to be possible to remove the fences and supplement the removed areas, and recognize the object, even if a part of the object is covered by fences. In this research, we aim to improve the accuracy of animal recognition using de-fencing on computer. (Keyword) Noice reduction / Image complement / Animal recognition (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI.2019.00114 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85080896317 (DOI: 10.1109/IIAI-AAI.2019.00114, Elsevier: Scopus)
20.	Kaoru Kimura, Felicia Misamanana Rabarison, Kawanaka Kohei, Hiroki Tanioka and Tetsushi Ueta : Animal Stamp Rally Application using Image Recognition, The 7th International Conference on Smart Computing and Artificial Intelligence (SCAI 2019), 1077-1078, Toyama, Jul. 2019. (Summary) Tokushima zoo is a municipal zoo housing a lot ofanimals for educational purpose. Thus, Tokushima city wantsvisitors to observe animals as many as possible whilst havingfun. The animal stamp rally is a game that visitors collect aseries of stamps at around animals which visitors have observed.Not to provide real stamps, tables and stamp cards, we developan animal stamp rally application. Thanks to the ConvolutionalNeural Network (CNN), the application can recognize stampsfrom snapshots of real animals instead of real stamps or QRcodes. Additionally, to improve animal recognition accuracy, theapplication uses GPS information. In this paper, we describesa progress of our development of the animal stamp rallyapplication, and estimates its recognition accuracy. (Keyword) Image Recognition / GPS Camera / Stamp Rally (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI.2019.00234 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85080922805 (DOI: 10.1109/IIAI-AAI.2019.00234, Elsevier: Scopus)
21.	Hiroki Tanioka : AITOK at the NTCIR-14 OpenLiveQ-2 Task, Proceedings of the 14th NTCIR Conference on Evaluation of Information Access Technologies, Tokyo, Jun. 2019. (Summary) The AITOK team participated in the NTCIR-14 OpenLiveQ2 Task. This report describes our approach to ranking question-answeringlists of Yahoo! Chiebukuro search results for the queries of Query IDs,and discusses our results of test results. Our approach intends to makesure of the degree of catchy to question-answering, thereby integrates twostrategies. The first strategy is statistics based approach with questionsand clickthrough data. The second strategy is natural language basedapproach with question-answer pairs. In the offline test, we struggled increasing Q-measure using these two strategies day by day. Additionally,we employed manually dynamic programming approach for optimizationof Q-measure. Although the approach is very simple, a result sorted bya score mainly based on page view is the best in all results. However, thebest result in the offline test is not enough in the online test. Instead, theother result sorted by a score based on clickthrough with last-updatedtime in descending order is ranked in the top group. (Keyword) Question-answering / information retrieval / Freshness
22.	Hiroki Tanioka, Kenji Matsuura, Stephen Githinji Karungaru, Naka Gotoda, Kai Tomohiro, Wada Tomohito and Takai Yohei : Player Tracking in Sports Video using 360 Degree Camera, IEEE International Conference on Computational Photography 2019, Tokyo, May 2019.
23.	Kimura Kaoru, Felicia Misa Rabarison, Kawanaka Kohei, Hiroki Tanioka and Tetsushi Ueta : Local Zoo Animal Stamp Rally Application using Image Recognition, IEEE International Conference on Computational Photography 2019, Tokyo, May 2019.
24.	Kohei Kawanaka, Felicia Misa Rabarison, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : Improving Animal Recognition Accuracy using Deep Learning, IEEE International Conference on Computational Photography 2019, Tokyo, May 2019.
25.	Stephen Githinji Karungaru, Kenji Matsuura, Hiroki Tanioka, Tomohito Wada and Naka Gotoda : Ground Sports Strategy Formulation and Assistance Technology Development - Player Data Acquisition from Drone Videos, Proceedings of 8th International Conference on Industrial Technology and Management, 322-326, Mar. 2019. (Summary) Sports analytics aims to provide a competitive advantage to a team or individual in sports by analyzing a collection of relevant statistics. Better strategies in play can be formulated using the information. The core technology necessary for the analysis is the player detection and tracking. Due to limitation of fixed cameras, this work presents a court detection using markers, player detection and tracking using a drone. The method presented is based on perspective transformation, template matching and particle filters. In this work, we can visualize the movement trajectory of a player and extract the real world (on court) player position irrespective of drone height and position. (Keyword) Sports Strategy / Assistance Technology / Drones (Link to Publication Site) ● Publication site (DOI): 10.1109/ICITM.2019.8710735 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85066608325 (DOI: 10.1109/ICITM.2019.8710735, Elsevier: Scopus)
26.	Naoya Kohda, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Tomohito Wada and Naka Gotoda : Technology-Supported Single Training for One-on-One in Basketball Matches, Proceedings of IEEE TALE2018, 447-453, Wollongong, Australia, Dec. 2018. (Link to Publication Site) ● Publication site (DOI): 10.1109/TALE.2018.8615318 (Link to Search Site for Scientific Articles) ● Search Scopus @ Elsevier (DOI): 10.1109/TALE.2018.8615318 (DOI: 10.1109/TALE.2018.8615318)
27.	Hiroki Tanioka, Kaoru Kimura, Kazuma Takaoka, Ryohei Nakatani and Yoshitaka Uchida : Automatic Generation of Japanese Question-Answering Pairs, Fourth Asia Pacific Corpus Linguistics Conference (APCLC 2018), Takamatsu, Sep. 2018. (Keyword) Question-Answer Corpus / Automatic Generation / machine learning / Evaluation Metrics
28.	Mamoru Sakai, Hiroki Tanioka, Kenji Matsuura, Masahiko Sano, Kenji Ohira, Tetsushi Ueta and Sakaguchi Hiroaki : Evaluating Hitting Skills of NPB Players with Logistic Regression Analysis, CSCE'18 Proceedings, 313-319, Las Vegas, Jul. 2018. (Summary) Luck or unluck is supposed to exist in baseball. In this paper, we consider unpredictable "lucky" or "un-lucky" cases which are occurred on the ground in order to find out authentic ability of batters. If all cases are observable, hit probabilities can be calculated. However, it is impossible to gather all the information on the ground. Therefore, "luck" is defined as influence of information that cannot be observed. To provide more appropriate evaluation for NPB 1 players, this paper proposes a predicted BABIP using Logistic Regression and describes the predicted BABIP reducing the influence of luck on the assessment of hitting ability. Analysis results show that there is a possibility of appropriate evaluation on hitting abilities of NPB players using the predicted BABIP. (Keyword) baseball / hitting / performance / evaluation / logistic regression
29.	Mamoru Sakai and Hiroki Tanioka : Keyword-based Challenges at the NTCIR-13 MedWeb, Proceedings of the 13th NTCIR Conference on Evaluation of Information Access Technologies, Dec. 2017. (Keyword) Keyword-based / Logistic Regression / Support Vector Machine
30.	Hiroki Tanioka : Response Generation for Grounding in Communication at NTCIR-13 STC Japanese Subtask, Proceedings of the 13th NTCIR Conference on Evaluation of Information Access Technologies, Dec. 2017. (Keyword) dialogue understanding / Common Ground / Support Vector Machine
31.	Hiroki Tanioka : Super Easy Way of Building Image Search with Keras, SIGIR 2017 LIARR Workshop, Aug. 2017. (Keyword) Image Search / Elasticsearch / Keras / Jupyter Notebook
32.	Hiroki Tanioka, Takafumi Bando, Kenji Matsuura, Kenji Ohira, Masahiko Sano and Tetsushi Ueta : Improving Document Availability in Storage, --- Where're our important documents for daily business operations? ---, The 9th International Conference on E-Service and Knowledge Management (ESKM 2017), 997-1000, Hamamatsu, Jul. 2017. (Summary) This paper provides a case study of improving availability for searching documents in an academic facility. Every day some documents are needed for our business and academic activities in the Center for Administration of Information Tech- nology (AIT) in Tokushima University. AIT has been employing and operating Information Security Management System (ISMS) since 2014. ISMS requires to maintain Confidentiality, Integrity and Availability. Particularly, AIT places importance on Avail- ability. However, time-consuming is a problem to find documents in the daily operation. A case study was conducted to put a full-text search engine in the facility, considering the objective and efficiency about finding documents in the file-server. In the process, it was found out that document freshness is important to academic office workers for their job. Then, the effectiveness of a score of TF-IDF with document freshness was assessed. Therefore, availability for finding documents was significantly improved using a full-text search engine with document freshness. (Keyword) Document management / information retrieval / Elasticsearch (Link to Publication Site) ● Publication site (DOI): 10.1109/IIAI-AAI.2017.139 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85040546313 (DOI: 10.1109/IIAI-AAI.2017.139, Elsevier: Scopus)
33.	Michitomo Morii, Hiroki Tanioka, Kenji Ohira, Masahiko Sano, Kenji Matsuura, Yosuke Seki and Tetsushi Ueta : Research on Integrated Authentication Using Passwordless Authentication Method, COMPSAC Workshops 2017 Student Research Symposium, 682-685, Torino, Jul. 2017. (Summary) Currently, authentication methods using ID and password are widely used and fulfilled central roles in various information systems and services. Our university also uses ID and password for authentication of most services. However, passwords have various problems such as reuse, phishing and leakage. This research is a practical experiment in order to implement an integrated authentication system without password. Shibboleth is introduced to our university, providing capabilities of web single sign-on and attribute exchange framework for organizational services. The Fast IDentity Online (FIDO) is adopted into Shibboleth as an external authentication, to realize passwordless authentication. Furthermore, we held a feasibility test of an integrated authentication system without password, and considered problems of the passwordless authentication method using FIDO. (Keyword) Authentication / Passwordless / Shibboleth / FIDO (Link to Publication Site) ● Publication site (DOI): 10.1109/COMPSAC.2017.198 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-85031916565 (DOI: 10.1109/COMPSAC.2017.198, Elsevier: Scopus)
34.	Wu Jiangning, Hiroki Tanioka, Wang Shizhu, Pan Donghua, Yamamoto Kenichi and Wang Zhongtuo : An improved VSM based information retrieval system and fuzzy query expansion, FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY, PT 1, PROCEEDINGS, Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science), Vol.3613, 537-546, Aug. 2005. (Link to Publication Site) ● Publication site (DOI): 10.1007/11539506_68 (Link to Search Site for Scientific Articles) ● Summary page in Scopus @ Elsevier: 2-s2.0-26944486877 (DOI: 10.1007/11539506_68, Elsevier: Scopus)

Proceeding of Domestic Conference:

1.	Hiroki Tanioka, Mitsuhiro Sato and Ryuji Tsuge : 小学生年代のサッカースクールにおける楽しさの分析, Japanese Society of Science and Football Congress 19, Mar. 2022. (Summary) スポーツ等において，身体活動を楽しむことや人との関わりの重要性が示唆されているが，どのような活動が楽しみにつながるのかについては明らかになっていない．そこで，サッカースクールに通うスクール生にアンケートを取り，スクールに参加してどのようなことを感じているのか，サッカースクールの活動を楽しめているのか，どんな活動が楽しさにつながるのかなどについて分析を行った． (Keyword) Soccer / Fun / Correlation analysis
2.	Takahiko Kojo, Masahiko Sano, Kenji Matsuura, Hiroki Tanioka and Tetsushi Ueta : Classification of Research Fields, Shikoku-Section Joint Convention Record of the Institutes of Electrical and Related Engineers, 210, Sep. 2021. (Summary) 科学研究費助成事業(以下，科研)のデータベース(KAKEN)[2]を基に研究組織の研究を分類し，科研と所属組織の各研究分野への注力度合いとその推移を可視化する場合において，分類基準が変更になった際の時系列的に連続した分類とする分類手法の提案と実装
3.	Hiroki Tanioka, Takafumi Bando, Kenji Matsuura, Tetsushi Ueta, Masahiko Sano and Kensuke Inai : Electronic Application System for Security Export Control Procedures, 国立大学法人情報系センター協議会 (NIPC), Sep. 2021. (Keyword) Security Export Control / Electronic Application / Online / Digital Transformation Promotion
4.	Hiroki Tanioka and Rina Yano : Online Workshop Practices using Breakout Rooms, The 46th Annual Conference of JSiSE, No.E3-3, 43-44, Sep. 2021. (Summary) COVID-19 の流行が収束せず，多くの授業がオンライン開催を余儀なくされている現状である. 本稿では，このような状況において，徳島大学の歯学部 1 年生向けの医療情報処理でワークショップ「しりとりゲーム」と「スパイゲーム」について，2019 年度は対面形式，2020 年度と 2021 年度はオンライン形式で実施したので，その目的，方法，内容，授業形式による比較，課題について述べる.特に，2021 年度は，ブレイクアウトルームを用いることで，オンライン形式でも対面形式に近い形で実施できたのでこれについて詳しく述べる. (Keyword) workshop / online lessons / breakout rooms / information education
5.	Kenji Matsuura, Naka Gotoda, Tomohito Wada and Hiroki Tanioka : Discussion of Learning Support for Team Tactics with 2D Simulator, The 46th Annual Conference of JSiSE, 45-46, Sep. 2021. (Summary) チームスポーツでは，チーム戦術を個々のプレイヤが理解・共有し，状況に応じて連携して実践適用することで，ゲームを優位に進めることができる．特に初学者においては，その概念形成をフィールドやコート全体の俯瞰により得ることが有用である．そこで，本研究では，チームスポーツのゲームを二次元俯瞰視するシミュレータを応用した学習支援環境を構築している．現状までの開発状況とその課題を述べ，今後の可能性を議論する．
6.	Hiroki Tanioka and Miyuki Morimoto : 相関関係による得点に繋がるプレーの分析, Japanese Society of Science and Football Congress 18, Mar. 2021. (Summary) 本稿では，J1 サッカーリーグ 2019 年のデータを対象に，勝点に結びつくプレーを相関分析によって明らかにしようとした.まず，勝点と強い相関を持つピタゴラス勝率，次にピタゴラス勝率を構成する得点と失点について，それぞれに高い相関を持つプレーデータと，そのプレーデータ間の関係を掘り下げて分析を行なった. (Keyword) Soccer / Correlation analysis / Tactical analysis
7.	Shu Yano, Kenji Matsuura, Hiroki Tanioka, Tomohito Wada and Naka Gotoda : Action selection support for basketball player holding a ball, The 45th Annual Conference of JSiSE, 121-122, Sep. 2020. (Summary) バスケットボールにおいて，各プレイヤはシュートに向けてその場の状況を判断し，行動を選択する．本研究では，ボール保持プレイヤに着目し，その行動選択の学習支援を行う．手法として，パス，ドリブル，シュートに焦点をあて，ボール保持プレイヤが置かれている状況下での行動選択肢を試合データから自動抽出する．さらに，その選択行動をシュートを目標として指標化する．自動抽出と指標の視覚化をするシステムの設計，構築を行う．
8.	Rina Yano, Hiroki Tanioka, Kenji Matsuura, Masahiko Sano and Tetsushi Ueta : Development of an online test aiming quantitatively measurement of computational thinking, The 45th Annual Conference of JSiSE, 43-44, Sep. 2020. (Summary) 2020 年度より小学校ではプログラミング教育が必修化されたが，学習者の能力の成長を定量的に測定する手法は確立されていない．本研究ではプログラミング教育によって養われるとされるプログラミング的思考には計算論的思考が含まれると考え，計算論的思考力を客観的・定量的に測定するためのテストを開発してきたが，小学校教育の現状を鑑みて遠隔授業等にも導入できるようオンライン形式で公開した．本稿では，事前調査として大学生 38 名を対象に行ったテスト結果について報告する． (Keyword) Programming Education / ICT / Learning Analytics
9.	小林崇弘, Masahiko Sano, Kenji Matsuura, Hiroki Tanioka, Kenji Ohira and Tetsushi Ueta : 希望研究室選択時の研究内容調査支援システム, 第82回全国大会講演論文集, No.4ZF-03, Mar. 2020.
10.	Shu Yano, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Naoya Kohda, Tomohito Wada and Naka Gotoda : A learning environment of a systematic and tactical style in team sports, The 44th Annual Conference of JSiSE, 213-214, Sep. 2019. (Summary) 集団対戦型フィールドスボーツにおいて，各プレイヤがチームの基本戦術を理解することが大切になる.したがって，各基本戦術を認識し，以後の展開予測までできることが望ましい.本研究では，バスケットボールにおける戦術の適用場面を自動抽出し，シミュレーションと実際の映像を同期させて見せることによって，戦術が適用される場面での判断の基準を形成できるシステムを構築する. (Keyword) 可視化 / チーム戦術 / バスケットボール / simulation
11.	Naoya Kohda, Shu Yano, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Tomohito Wada and Naka Gotoda : Improvement of software defense design in feint learning, The 44th Annual Conference of JSiSE, 169-170, Sep. 2019. (Summary) 本研究では，バスケットボールの 1 対 1 の場面におけるフェイント動作について取り上げる. その学習支援として，学習者にソフトウェアで実装したディフェンスを提示する.このディフェンスを突破する練習を行う環境の設計に関する改善を図る.フェイント動作を身体部位の動作の仕方に着目し分析する.そのために事前に撮影した動画を解析し，学習者ごとのモデル化を行い，それに対応するディフェンスの挙動を提案する. (Keyword) 学習支援 / 支援モデル / 統合運動 / フェイント動作
12.	Hiroki Tanioka, Rina Yano, Kenji Matsuura, Masahiko Sano and Tetsushi Ueta : A Workshop Based Programming Educational Practice, --- Welcome to CoderDojo Tokushima ---, The 44th Annual Conference of JSiSE, 195-196, Sep. 2019. (Summary) CoderDojo Tokushima は 2017 年 12 月から継続的に開催されている.CoderDojo は小学生~高校生がプログラミングを学習する場としてアイルランドで生まれ，110 ヶ国，国内は 169 ヶ所で開催されている.参加者から参加費を徴収しないことを除いて，内容については主催者に委ねられているため，我々は，参加者の自主的な学習や情報交換の促進を重視し，ワークショップ形式を用いている.毎回のアンケートの統計情報からその効果と課題について述べる. (Keyword) workshop / Programming Education / CoderDojo / Questionnaire Analysis
13.	Hiroki Tanioka : ビッグデータで勝機を見いだす!スポーツデータ分析, 情報処理学会第81回全国大会, Mar. 2019.
14.	平松孝雄, Masahiko Sano, Kenji Matsuura, Hiroki Tanioka, Kenji Ohira and Tetsushi Ueta : ゼミ議事録と研究成果物の蓄積と利用の支援システムの開発, 第81回全国大会講演論文集, No.5ZH-05, Mar. 2019.
15.	小林崇弘, Masahiko Sano, Kenji Matsuura, Hiroki Tanioka, Kenji Ohira and Tetsushi Ueta : 研究室選択時の意思決定支援システム, 第81回全国大会講演論文集, No.5ZH-09, Mar. 2019. (Summary) 多くの大学では研究室調査を学生が各自で行い，学生が行きたいと思う研究室を選択する．本研究室ではWebサービスを用いて入力されたキーワードに関する論文を本学が保有する研究者情報DBより取得し，利用者に提示することによる研究室の配属の支援を行っている．しかし，先行研究では幾つかの課題があり，その一つに日本語キーワードから英語論文が検索できない問題点があった．本研究では単語翻訳機能を用いて関連する英語論文を提示する機能を追加する．また，先行研究の有用性の追跡調査を行い評価する
16.	Hiroki Tanioka, Nakatani Ryohei and Uchida Yoshitaka : 統計情報を用いた人の心を捉える質問応答文の分析, 言語処理学会第25回年次大会発表論文集, 866-867, Mar. 2019. (Keyword) 質問応答 / information retrieval / accuracy
17.	Nishimura Yuki, Masahiko Sano, Kenji Matsuura, Hiroki Tanioka, Kenji Ohira and Tetsushi Ueta : A Management Method of Research Results Metadata, Journal of Shikoku-Section Joint Convention of the Institutes of Electrical and Related Engineers, 232, Sep. 2018. (Summary) 研究活動における論文や実験データなどの研究成果物を蓄積と再利用を支援するため，CMSの一つであるMaharaを機能拡張することを目的とする．研究成果物情報をRDF化してDBに集積することにより，研究情報成果物に対するメタ情報の効率的な利活用の管理を支援する．
18.	Hiramatsu Takao, Masahiko Sano, Kenji Matsuura, Kenji Ohira, Hiroki Tanioka and Tetsushi Ueta : A Management System for Reusing Meeting and Research Results, Journal of Shikoku-Section Joint Convention of the Institutes of Electrical and Related Engineers, 233, Sep. 2018. (Summary) 大学研究室での学生のゼミ活動でにおいて，その議論内容の時間経過による忘失を防止するため，テキストや映像・音声データなどによりゼミ議事録として記録される．本研究では，ゼミ議事録と研究成果物の参照や再利用を活性化し研究活動の効率化を支援するため，ゼミ議事録にメタデータを付与することで機械的に整理して管理し，メタデータを利用した情報検索を行うシステムの開発を目的とする．
19.	Shu Yano, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Tomohito Wada and Naka Gotoda : Fundamental discussion for understanding tactics of team-match sports, Journal of Shikoku-Section Joint Convention of the Institutes of Electrical and Related Engineers, 215, Sep. 2018.
20.	幸田尚也, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, 和田智仁 and Naka Gotoda : Breakthrough skill development for one-on-one match in competitive sports, 教育システム情報学会第43回全国大会講演論文集, 83-84, Sep. 2018.
21.	Takuya Ishioka, Naka Gotoda, 米谷雄介, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, 和田智仁, 國枝孝之, Rihito Yaegashi and Toshihiro Hayashi : Development of Sports Support System Using Superimposed Movie with Reflection Materials, 教育システム情報学会第43回全国大会講演論文集, 77-78, Sep. 2018.
22.	Hiroki Tanioka, Ryohei Nakatani, Suzuko Yogi and Yoshitaka Uchida : 質問応答システムの基盤化と応答可能性の判定, 言語処理学会第24回年次大会発表論文集, 651-653, Mar. 2018. (Summary) ビジネスシーンを想定した業務システムとして，Chatbot や質問応答システムの構築を検討する際には，応答不可能な質問に対する適切に対応することが求められる．しかしながら，質問応答システムが，ユーザからの質問を分析し，質問に対して適切に応答できるか否か (基盤化が必要か否か) を判定する手法については，未解決であった．ユーザの質問文を解析して形態素及び単語分散表現で表し，サポートベクトルマシン (Support Vector Machine; SVM) を用いて自動分類する手法を提案する． (Keyword) Question Answering System / interactive interface / Common Grounding
23.	Hiroki Tanioka, Kenji Matsuura, Tetsushi Ueta and Fumiaki Kawano : BYOD環境によるワークショップ型実習の試みとその課題, 大学教育カンファレンス in 徳島, Jan. 2018. (Summary) 2017年度前期に，徳島大学歯学部1年生を対象とした実習ベースの授業科目「医療情報処理」全16回の講義の11回分の講義と実習及び成績評価までを担当し，いくつかの知見が得られたので，その結果と課題について報告する．本講義は，受講者全員(歯学科40名，口腔衛生学科15名)が個人PCを持ち込むワークショップ形式1) の講義であったため，BYOD環境を全学展開する上での試金石となったので，これもあわせて報告する． (Keyword) BYOD / workshop
24.	Ryo Takeno, Masahiko Sano, Kenji Matsuura, Hiroki Tanioka, Kenji Ohira, Yosuke Seki and Tetsushi Ueta : Visualization of Research Field in Research Unit, AXIES2017, Dec. 2017. (Summary) 研究組織において，自組織及び学界での研究の動向を把握しておくことは重要である．本研究では，科学研究費助成事業による分野分類を用いて研究組織の研究分野を推定し，各分野における注力度合いを可視化する．
25.	Shin Maruyama, Kenji Ohira, Masahiko Sano, Hiroki Tanioka, Kenji Matsuura and Tetsushi Ueta : Management solution of PC classroom to balance unification of desktops and terminal's speed, IPSJ SIG Technical Report, Vol.2017-IOT-39, No.14, 1-6, Sep. 2017.
26.	西村友基, Masahiko Sano, Kenji Matsuura, Hiroki Tanioka, Kenji Ohira, Yosuke Seki and Tetsushi Ueta : A metadata management method of arifacts in Mahara, Forum on Information Technology, 349-351, Sep. 2017. (Summary) 大学における研究活動では，論文や発表スライド，実験データなどの研究成果物が生成・収集される．収集データを利活用する際には，成果物内の情報あるいは成果物に付与されるメタ情報を検索に用いる．本研究では，付与されたメタ情報同士の関係性を解析し，利用者に提示することにより，研究活動の効率化を目的とする．本論文では，eポートフォリオであるMaharaを対象としてメタ情報管理のための手法を提案する．メタ情報はメタ情報記述のための枠組みであるRDFに基づき表現することで，意味に基づいた検索や，他の公開されているメタ情報との連携を可能とする．
27.	Tomohiro Takeda, Kenji Ohira, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : A Rogue Access Point Detection Method with Estimation and Trace of Internet Reachability Check Packets, 情報処理学会全国大会講演論文集, 3-161-3-162, Mar. 2017.
28.	Izumi Nishina, Kenji Ohira, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : A Security Warning System Associating Detected Actions with their Risks to Motivate Users to Take Countermeasures, 情報処理学会全国大会講演論文集, 3-601-3-602, Mar. 2017.
29.	Hirokazu Nakano, Kenji Ohira, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : A method of two-factor authentication with a student ID card and a password to WLAN client authentication, 情報処理学会全国大会講演論文集, 3-573-3-574, Mar. 2017.
30.	Yamamoto Kenichi, Hiroki Tanioka and Tonoi Kayoko : 動向情報の検索による情報編纂, Proceedings of the Annual Conference of JSAI, Vol.7, 3H93, 2007. (Summary) <p>近年，電子化された情報の増加と検索技術の発展により，比較的単純な情報であれば検索技術を用いて容易に得ることが可能となってきた．しかし，例えばある会社の株価の変動と同期している株価の変動をもつ会社や，ある製品の売上の増加と共に使用されるようになった単語を知りたい場合など，入手した統計情報を検索者の検索意図に沿った形に加工し，さらにその統計情報を利用して再検索するといった情報編纂はまだ難しい．そこで我々は，株価などの統計情報に加えて単語の出現傾向も1つの動向情報と見なし，単語と株価，単語と売上，株価と売上など，多様な動向情報間の類似性を算出することにより，これらの課題を解決する方法を提案する．</p> (Link to Publication Site) ● Publication site (DOI): 10.11517/pjsai.JSAI07.0_3H93 (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1390282763027049984 ● Search Scopus @ Elsevier (DOI): 10.11517/pjsai.JSAI07.0_3H93 (DOI: 10.11517/pjsai.JSAI07.0_3H93, CiNii: 1390282763027049984)

Products:

1.	Hiroki Tanioka : Japanese Question-Answering Corpus (JQAC), Jan. 2020. (Keyword) Question-Answering / corpus / Japanese

Et cetera, Workshop:

1.	Kanji Kawahara, Hiroki Tanioka, Tetsushi Ueta and Masato Ueguri : NPBとMLBにおける打者の抑え方の比較, IPSJ SIG Technical Report, Vol.2024-SI-1, No.36, 1-2, Jun. 2024. (Summary) 野球はスコアを競うスポーツであり，その結果は投手の能力に大きく左右される．投手が打者に打たれなければ，試合に負けることはない．しかしながら，1 人の投手が全試合に登板することはできず，中 4 日なら少なくともチームごとに 5 名の先発投手が必要となる．近年，NPB (Nippon Professional Baseball) で活躍した投手が MLB (Major League Baseball) へ移籍するケースが増加してきた背景としては，このような事情があると考えられる．そのような状況で，NPB で活躍した投手が MLB でも活躍するケースが増加している．一方，NPB で活躍していた投手が，MLB では活躍できていないケースもあることが，過去のインタビューや分析から明らかになってきた．NPB と MLB では投球戦略やプレースタイルに違いが影響しているという．そこで本研究では，NPB と MLB における打者の抑え方の違いを明らかにすることを目的とし，NPB と MLB のそれぞれの対戦成績ついて，球速などの投球内容と，その結果である被打と失点との関係について，Spearman の順位相関で相関分析で分析した結果，NPB も MLB でも共通して，1 打席の球数や前の球との縦座標の差の最小値と被打，塁にランナーがいることと失点の相関が高いことが分かった．また，NPB ではストライクゾーンに投げた割合と被打との間に正の相関がみられたが，MLB ではストライクのカウント数と被打との間に負の相関がみられた． (Keyword) correlation function / Baseball / Pitching
2.	Hiroki Tanioka, Boyuan Wang and Masahiko Sano : Keyword search for transcriptions containing recognition errors, IPSJ SIG Technical Report, Vol.2023-IFAT-150, No.2, 1-2, Mar. 2023. (Summary) In recent years, speech recognition technology using deep learning has been developing remarkably, and automatic transcription of minutes by AI is increasingly being used to automate the creation of minutes of online meetings and other events. After a meeting, the minutes automatically transcribed by AI should be stored in a database without manual correction as much as possible, so that the information can be viewed by keyword search and other means. Storing the transcribed text of a meeting without correction has the advantages of not requiring large storage space and allowing users to check the content relatively quickly even if it contains errors. Therefore, we have developed a system that allows users to browse a database of error-containing texts by keyword search. The problem here is that the minutes transcribed by the speech recognition function may contain errors, and therefore cannot be searched by keyword search using text strings. Based on the hypothesis that the erroneous character strings from speech recognition are likely to be similar to the phoneme vectors of the spoken words, we propose to perform keyword searches using the edit distance of the phoneme vectors. Experimental results of keyword retrieval on a corpus with transcription confirmed that both reproduction rate and fit rate were improved. (Keyword) transcription / evaluation / information retrieval / phoneme representation / edit distance
3.	Hiroki Tanioka and Kenji Taniwaki : Program Information Extraction Using Gradient Boosting, IEICE Technical Report, Vol.121, No.415, NLC2021-37, 54-55, Mar. 2022. (Summary) Although video distribution services using the Internet have been launched one after another, the authors currently perform most of the work of registering program information recommended by television production offices into a unified database system by hand, which be expected to become difficult to cope with the increase in the number of programs in the future. Therefore, this study aims to automate the process of converting program information into metadata. Specifically, text data and image data are extracted from PDF and other document data and converted into program metadata such as program titles and program contents. We have confirmed that the metadata of program information can be converted to metadata with high accuracy by learning and estimating the metadata using gradient boosting, using manually generated program information as training data. (Keyword) text mining / Gradient boosting / Keyword extractor / Metadata
4.	Ryota Kohri and Hiroki Tanioka : Behavioral analysis that contributes to improving the winning percentage of the werewolf team in the werewolf game, IPSJ SIG Technical Report, Vol.2021-ICS-202, No.3, 1-4, Mar. 2021. (Summary) In recent years, the number of systems that utilize artificial intelligence is increasing. Even in Shogi and Go, which are perfect information games, artificial intelligence has been used to beat masters. Furthermore, artificial intelligence is being actively researched in Poker and Mahjong, which are imperfect information games. On the other hand, the werewolf game focused in this study is an incomplete information collective game, and it is necessary to aim for winning as a team using incomplete information. In this paper, we report that artificial intelligence improves the winning percentage by taking a team strategy in a 15-player werewolf game. (Keyword) werewolf game / team strategy / artificial intelligence
5.	Hiroki Tanioka : Practical Plan for AI / Data Science Education with a Focus on Social Implementation, JSiSE Resaeach Report 2020 Special Issue Articles Conference, Mar. 2021. (Summary) In order to transform a new society, there is an urgent need to develop human resources who have acquired knowledge and skills related to AI / data science, and there is an urgent need to build an education system related to this. AI / data has the ability to design social implementations. There is a need for educational methods and systems that can develop human resources who can carry out using all knowledge and skills including science. Therefore, we decided to open a cram school (open lecture) for practicing AI / data science education for elementary school students to high school students. In this paper, we describe the curriculum proposal, the correspondence with AI 5 Big Idea, the questionnaire for measuring the effect, the breakdown of the participants, and the future prospects. (Keyword) artificial intelligence / data science / social Implementation
6.	Renpei Yamamoto, Kenji Matsuura, Hiroki Tanioka, Tomohito Wada and Naka Gotoda : Gaze learning environment for defensive players in one-on-one match situation of basketball, JSiSE Student Reseach Report in Shikoku, 231-232, Mar. 2021.
7.	Kodai Nagataki, Kenji Matsuura, Hiroki Tanioka, Tomohito Wada and Naka Gotoda : Model-view learning environment for tactics training of basketball, JSiSE Resaeach Report, 229-230, Mar. 2021.
8.	Hiroki Tanioka, Taisuke Suzuki, Hiroshi Imai, Shikanai Hiroshi, Tomoyuki Noguchi, Yumi Hattori, Shota Mochizuki and Toshitaka Matsuda : Keyword analysis using the annotation function in the Daily Report Entry System, IEICE Technical Report, Vol.120, No.374, 19-22, Feb. 2021. (Summary) In Japanese organizations, it is not uncommon for class members to submit reports in the form of daily reports (weekly or monthly reports in some cases) for the purpose of reporting the progress of work. With the development of ICT, the number of systems and applications for easily inputting daily reports has increased, but it is difficult to say that the input daily reports are being used efficiently. By clarifying which keywords in the reported daily report are notable and why they are attracting attention, the authors can motivate them by giving feedback to the reporter and give the evaluators. We have developed an annotation function and a badge function, considering that it will lead to accurate evaluation by reporting. We also report an attempt to extract and visualize the keywords of interest using the tf-idf of words contained in the annotated character string and distributed expressions. (Keyword) text mining / Annotation / Keyword extract
9.	Rina Yano, Hiroki Tanioka, Kenji Matsuura, Masahiko Sano and Tetsushi Ueta : Quantitative Measurement and Analysis to Thinking as a Way of Programming for Elementary School in Japan, IPSJ SIG Technical Report, Vol.2020-CLE-30, Mar. 2020. (Summary) Programming education will be compulsory in elementary schools in Japan from 2020, but thereis still little information on how to measure the effects of lessons conducted for students. In this study, inorder to objectively and quantitatively measure the educational effect of programming education, we developed a test of measuring an ability of thinking as a way of programming. In this paper, we evaluate themeasurability of the test using the Fishers exact test for elementary school students answers, focusing onthe grade, the programming experience, and the difference between the results before and after the workshop. (Keyword) Programming Education / evaluation / analysis
10.	Katsumi Oshiba, Naoya Kohda, Kenji Matsuura, Hiroki Tanioka, Naka Gotoda and Tomohito Wada : Development and Evaluation of Software Offence in Basketball, JSiSE Resaeach Report, 203-204, Mar. 2020.
11.	Naoya Kohda, Katsumi Oshiba, Kenji Matsuura, Hiroki Tanioka, Naka Gotoda and Tomohito Wada : Learning Environment for Feint Motion by Software Defense Using Probability, JSiSE Resaeach Report, 201-202, Mar. 2020.
12.	Sakai Mamoru, Hiroki Tanioka, Kenji Matsuura, Naka Gotoda and Tomohito Wada : Proposal of Batting Ability Evaluation Index using Logistic Regression Analysis, IPSJ SIG Technical Report, Vol.2020-MPS-127, No.8, 1-6, Mar. 2020. (Summary) In baseball, BABIP (Batting Average on Balls In Play) which is the percentage of hits that hitthe ground excluding the home run is used as an index to judge the batter's ability.A hit ball is affectedby various factors, so it is difficult to think that the result of a hit or out represents only the ability of thepitcher and batter. In this study, the OPPB(On-base plus predicted BABIP) based on the logistic regressionmodel is calculated using the properties of the hit ball, the coordinates of the hit ball, the defensive strengthof the opponent team, the batter's running ability, and so on. (Keyword) baseball / evaluation / logistic regression
13.	Takuya Ishioka, Naka Gotoda, Ryota Akagi, Yuki Hirasawa, Kenji Matsuura, Hiroki Tanioka, Stephen Githinji Karungaru, Tomohito Wada, Yusuke Kometani, Takayuki Kunieda, Rihito Yaegashi and Toshihiro Hayashi : Evaluation of the Video System for Motor Skill Training Superimposing Symbols, IEICE Technical Report, Vol.ET2018-111, 135-140, Mar. 2019.
14.	Kawanaka Kohei, Rabarison Misamanana Felicia, Hiroki Tanioka, Kenji Ohira, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : Improving Animal Recognition Accuracy using De-fencing, IPSJ SIG Technical Report, Vol.2019-ICS-194, No.9, 1-6, Mar. 2019. (Summary) When a computer recognizes an animal on an image, an accuracy of animal recognition decreases if a part of the animal's body is covered by fences. On the other hand, when a human recognizes a object on the image, it is known to be possible to remove and supplement the fences, and recognize the object, even if a part of the object is covered by fences. In this research, we aim to improve the accuracy of animal recognition using de-fencing on computer. (Keyword) Noice reduction / Image complement / Animal recognition
15.	Felicia Misamanana Rabarison, Kohei Kawanaka, Hiroki Tanioka and Tetsushi Ueta : Local Zoo Guide Application using Image Recognition, IEICE Technical Report, Vol.PRMU2018-118, No.459, 21-26, Feb. 2019. (Summary) Most of the information for guide in the Tokushima zoo is in Japanese. However, it is difficult to understand Japanese guide for foreign tourists. Therefore, we need to develop an animal guide application or robot without language query for foreign visitors. One of the ideas is that Convolutional Neural Network (CNN) is employed for image recognition for making image query on the application. The application is available to display instantly information about animals in an image or video. For our experiment, pictures including 30 kinds of animals are collected. Around 4,000 pictures were taken from Tokushima zoo for the purpose of this experiments. (Keyword) Convolutional Neural Network / Image Recognition / VGG-16 / Fine-tuning
16.	Hiroki Tanioka, Kenji Matsuura, Tetsushi Ueta and Fumiaki Kawano : Report for Learning Email Etiquette in Group, JSiSE Resaeach Report, Vol.33, No.4, 21-22, Oct. 2018.
17.	Kenji Matsuura, Hiroki Tanioka and Naka Gotoda : Modeling issue of experience, prediction, perturbation, and coordination in motor-skill development, JSiSE Resaeach Report, Vol.33, No.4, 1-6, Oct. 2018.
18.	Mamoru Sakai, Hiroki Tanioka, Kenji Matsuura, Masahiko Sano, Kenji Ohira, Tetsushi Ueta and Sakaguchi Hiroaki : Evaluating Hitting Skills of NPB Players with Logistic Regression Analysis, IPSJ SIG Technical Report, Vol.2018-MPS-119, No.8, Jul. 2018. (Summary) When we watch a baseball game, we often witness scenes that a line drive flies out in front of a fielder. On the other hand, a pop fly falls between infielders and outfielders. In such a scene, we viewers express unlucky or lucky for the batter. Therefore, luck is supposed to be existing in baseball. In other words, it is hard to say that hitting records depend on only the ability of the batters. In this paper, lucky or unlucky is occurred in unpredictable cases with human eyes. Hence, we consider that luck exists in a phenomenon beyond observers expected based on their observable information. (Keyword) baseball / hitting / performance / evaluation / logistic regression
19.	Takao Hiramatsu, Masahiko Sano, Kenji Matsuura, Kenji Ohira, Hiroki Tanioka and Tetsushi Ueta : A Meeting Recording System Concidered for Utilization, IPSJ SIG Technical Report, Vol.2018-CLE-24, No.17, 1-4, Mar. 2018. (Summary) 蓄積されたゼミ議事録や研究成果物が有効に利活用されていない問題に対して，それらの利活用や関連情報の再検索における手間の低減を目的とした手法を提案する．提案手法では，ゼミ議事録にメタデータを付与しながらの作成と，関連研究成果物との連携により，利活用する際の手間の低減を図る．本稿では，その提案手法とその実験システムの開発状況について述べる．
20.	Kana Okada, Masahiko Sano, Kenji Matsuura, Kenji Ohira, Hiroki Tanioka and Tetsushi Ueta : Support System for Laboratory Selection using Researchers Database, IPSJ SIG Technical Report, Vol.2018-CLE-24, No.7, 1-6, Mar. 2018. (Summary) 大学生の研究室配属前の希望度と配属後の満足度のミスマッチが問題となっている．本研究では，事前調査の結果，その理由の一つが希望研究室の調査不足であることに着目し，学生の調査不足を大学が所有する研究者情報データベースを活用して支援する希望研究室選択支援手法について提案する．本稿では，その提案手法とその実験システムを著者らが所属する組織にて評価した結果について述べる．
21.	Takuma Horiuchi, Kenji Ohira, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : スマートデバイス向けサードパーティーアプリの危険性検証ツールの提案, IEICE Technical Report, Vol.IA2017, No.75, 1-6, Mar. 2018.
22.	Takahiro Oriishi, Kenji Ohira, Michitomo Morii, Hiroki Tanioka, Masahiko Sano, Kenji Matsuura and Tetsushi Ueta : FIDOにおけるオーセンティケータ登録のパスワードフリーな運用に関する設計と実装, IEICE Technical Report, Vol.IA2017, No.74, 1-6, Mar. 2018.
23.	Kenji Ohira, Hirokazu Nakano, Hiroki Tanioka, Kenji Matsuura, Masahiko Sano and Tetsushi Ueta : A Study on Application of Two-Factor Authentication to Campus WLAN Connections, IPSJ SIG Technical Report, Vol.2017-IOT-38, No.3, 1-6, Jun. 2017.
24.	Michitomo Morii, Hiroki Tanioka, Kenji Ohira, Masahiko Sano, Kenji Matsuura, Yosuke Seki and Tetsushi Ueta : Research on authentication cooperation using passwordless authentication method, IPSJ SIG Technical Report, Vol.2017-IOT-37, No.8, 1-6, May 2017. (Keyword) Shibboleth / FIDO / Authentication / Password
25.	Kenji Matsuura, Masahiko Sano, Kenji Ohira, Hiroki Tanioka, Tetsushi Ueta and Yuki Nishimura : Descussion on Management and Application of Daily Academic Activity at a University, IPSJ SIG Technical Report, Vol.2016-CLE-20, Nov. 2016.
26.	Hiroki Tanioka, TAKASHI NAKAGAWA and MINORU MARUYAMA : An Improved Bayesian Filter for Spam Filtering, IPSJ SIG Notes, Vol.67, 73-76, Dec. 2007. (Summary) We propose an improved baysian filter for spam mail detection. Bayesian filter was used on existing learning spam filters which achieved some positive results. Although we expect them to improve accuracy while keeping the false positive rate low. Therefore, it is based on a thorough review of distribution for each word and mail that our means of spam mail detection shows an impressively higher accuracy than ever. (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1520290884505780736 (CiNii: 1520290884505780736)
27.	Takashi NAKAGAWA, Kayoko TONOI and Hiroki Tanioka : Evaluation of Trainable Spam Filters with a New Sensory Accuracy Measure, IEICE Technical Report, Vol.105, No.650, 37-41, Mar. 2006. (Summary) Total error rate is a novel accuracy measure to evaluate spam filters more precisely. This measure is the integration of existing static or dynamic accuracy measures such as an additional learning figure, a learning convergence property, recall and so on. This measure can evaluate or tune spam filters automatically. An actual tuning of a spam filter proved usefulness of this measure. (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1520290883931456640 (CiNii: 1520290883931456640)
28.	Hiroki Tanioka, Kayoko TONOI and Takashi NAKAGAWA : Evaluation Methods for Improvement of Sensory Accuracy in Spam Filter, IEICE Technical Report, Vol.105, No.529, 29-33, Jan. 2006. (Summary) We propose evaluation methods for improvement of sensory accuracy in spam mail filter. In spam mail filter, it is difficult to keep up with growing spam mail types, since there is real need for high-accuracy learning spam filter. Though we have no way of telling a classification performance by static accuracy, so we should estimate a dynamic accuracy in order to achieve a usability effect. The dynamic accuracy is divided into 3 evaluation figures, a primary figure, an additional learning figure and a learning convergence property. (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1520572357729624064 (CiNii: 1520572357729624064)
29.	Hiroki Tanioka and Minoru MARUYAMA : Ascii Art Pattern Recognition using SVM based on Morphological Analysis, Technical report of IEICE. PRMU, Vol.104, No.670, 25-30, Feb. 2005. (Summary) Late years, ASCII Art is commonly found among Internet-based E-mail and WWW, such as Face letters and pictographic characters have no linguistic characters. Normally, ASCII Art conveys the meaning of the word as visual meaningful information, but takes on a meaning of no lingistic. On this account, we are hard to understand the right meaning of the word. The most important thing about this study is a hight-accuracy ASCII Art pattern classification to extract the ASCII Art from the sentence using Support Vector Machine. (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1520009408953850624 (CiNii: 1520009408953850624)
30.	Hiroki Tanioka and Minoru MARUYAMA : Ascii Art Pattern Recognition using SVM based on Morphological Analysis, IEICE technical report. Natural language understanding and models of communication, Vol.104, No.668, 25-30, Feb. 2005. (Summary) Late years, ASCII Art is commonly found among Internet-based E-mail and WWW, such as Face letters and pictographic characters have no linguistic characters. Normally, ASCII Art conveys the meaning of the word as visual meaningful information, but takes on a meaning of no lingistic. On this account, we are hard to understand the right meaning of the word. The most important thing about this study is a hight-accuracy ASCII Art pattern classification to extract the ASCII Art from the sentence using Support Vector Machine. (Link to Search Site for Scientific Articles) ● CiNii @ National Institute of Informatics (CRID): 1520009408953850624 (CiNii: 1520009408953850624)

Report:

1.	Kanji Kawahara and Hiroki Tanioka : 打者を抑えるために必要な投手の特徴量についての研究, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.10, 70-73, Tokyo, Mar. 2024. (Keyword) Correlation coefficient / Baseball / Pitching
2.	Hiroki Tanioka and Kanji Kawahara : ベイズ推定を用いた得点力の評価手法, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.9, 139-140, Tokyo, Mar. 2023. (Keyword) Correlation coefficient / Soccer / Bayesian estimation
3.	Hiroki Tanioka and 中村颯己 : 相関係数を用いたNPBにおける球速についての分析, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.8, Tokyo, Mar. 2021. (Keyword) Correlation coefficient / Baseball / Ball speed
4.	Hiroki Tanioka and Ryota Kohri : 回帰分析を用いたサッカー J1 リーグにおける攻撃力と守備力の分析, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.8, Tokyo, Mar. 2021. (Keyword) Regression Analysis / Soccer / Pythagorean Expectation
5.	Sakai Mamoru and Hiroki Tanioka : ロジスティック回帰分析を用いた野球選手の打撃能力評価, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.7, Tokyo, Mar. 2020. (Keyword) Logistic Regression / Baseball / prediction
6.	Hiroki Tanioka : 徳島県の人口移動調査のためのトピック分析, 徳島県 EBPM推進事業研究報告書, Sep. 2019.
7.	Sakai Mamoru, Hiroki Tanioka and Shu Yano : 走力と安打確率の関係性 - ロジスティック回帰分析を用いた安打確率の予測, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.6, Tokyo, Mar. 2019. (Keyword) Logistic Regression / Baseball / prediction
8.	Hiroki Tanioka, Kimura Kaoru and Shu Yano : ロジスティック回帰分析を用いた得点チャンスの分析, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.6, Tokyo, Mar. 2019. (Keyword) Logistic Regression / Soccer / prediction
9.	Mamoru Sakai and Hiroki Tanioka : 野球における運とはなにか - ロジスティック回帰分析を用いた安打確率の予測, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.5, Tokyo, Mar. 2018. (Summary) 運の要素を排除して選手をより正しく評価するためには，各打球の安打確率を求め，選手の理論値としての予測打率を算出し，実際の打率と比較することで，運の影響を控除し，各選手の適正な評価を与えることが可能となると考える． (Keyword) Logistic Regression / Baseball / Batting Average / prediction
10.	Michitomo Morii and Hiroki Tanioka : 得点できる攻撃パターンは? - ディープラーニングを用いた得点予測, Research on Sports Data Analysis : Theory, Methodology, and Applications, Vol.5, Tokyo, Mar. 2018. (Summary) Deep Learning を用いて,どのようなパスコース(軌跡)がゴールしたかを分析し,その特徴を明らかにすることを目的とする. (Keyword) Deep Learning / Soccer / prediction

Patent:

1.	Toshitaka Matsuda and Hiroki Tanioka : 日報管理装置およびコンピュータプログラム, 2020-189182 (Nov. 2020), 2022-078483 (May 2020), 7394294 (Nov. 2023).

Grants-in-Aid for Scientific Research (KAKEN Grants Database @ NII.ac.jp)

Visualization system for performance analysis with multi-object tracking (Project/Area Number: 22K12293 )
Construction of a Synergy Effect Exploration Method and Its Application to Medical Information Databases (Project/Area Number: 20H05798 )
持続的研究の効率化を支援するシステムの研究 (Project/Area Number: 18K11572 )
Cyberphysical Space Connecting Local and Global Perspectives in Supporting Motor-Skill Development (Project/Area Number: 18H03344 )
Study on portfolio to support sustainable growth of research community (Project/Area Number: 15K12168 )
Search by Researcher Number (90785106)

Associate Professor : Tanioka, Hiroki

Research

Personal Web Page

Field of Study

Subject of Study

Book / Paper

Book:

Academic Paper (Judged Full Paper):

Academic Paper (Unrefereed Paper):

Academic Letter:

Review, Commentary:

Proceeding of International Conference:

Proceeding of Domestic Conference:

Products:

Et cetera, Workshop:

Report:

Patent:

Grants-in-Aid for Scientific Research (KAKEN Grants Database @ NII.ac.jp)