Yuichi SEI
| Department of Informatics | Professor |
| Cluster I (Informatics and Computer Engineering) | Professor |
| Department of Mechanical and Intelligent Systems Engineering | Professor |
| Cluster II (Emerging Multi-interdisciplinary Engineering) | Professor |
- Profile:
Specializing in the fusion of machine learning and software engineering, my research primarily focuses on the implementation of machine learning techniques to complex physical phenomena such as heat transfer and river systems, as well as in the domain of privacy-preserving data mining.
Researcher Information
Field Of Study
- Informatics, Intelligent informatics
- Informatics, Web and service informatics
- Informatics, Information security
- Manufacturing technology (mechanical, electrical/electronic, chemical engineering), Thermal engineering
- Social infrastructure (civil Engineering, architecture, disaster prevention), Hydroengineering
Career
- Apr. 2024 - Present
National Institute of Informatics, Visiting Professor - Oct. 2023 - Present
Mitsubishi Research Institute, Inc., Senior Fellow - Apr. 2023 - Present
The University of Electro-Communications, Graduate School of Informatics and Engineering, Professor - Apr. 2022 - Present
Waseda University, Sustainable Energy & Environmental Society Open Innovation Research Organization, Adjunct Researcher - Jan. 2018 - Mar. 2023
The University of Electro-Communications, Department of Informatics, Graduate School of Informatics and Engineering, Associate Professor - Feb. 2013 - Sep. 2022
Mitsubishi Research Institute, Inc., Visiting Researcher - Oct. 2019 - Mar. 2022
Japan Society and Technology Agency (JST), PRESTO Researcher - Mar. 2018 - Mar. 2022
Waseda University, Interdisciplinary institute for thermal energy conversion engineering and mathematics, Adjunct Researcher - Jan. 2013 - Dec. 2017
The University of Electro-Communications, Assitant Professor - Apr. 2009 - Dec. 2012
Mitsubishi Research Institute, Inc., Researcher
Member History
- Aug. 2024 - Present
Advisory Committee Chair, International Conference on Health Informatics, Intelligent Systems and Networking Technologies, Society - Apr. 2023 - Present
委員, データ合成技術評価委員会, Society - Feb. 2023 - Present
Treasurer, IEEE Computer Society Tokyo/Japan Joint Chapter, Society - Apr. 2022 - Present
運営委員, 情報処理学会 知能システム研究運営委員会, Society - 2022 - Present
Program Committee Board, International Joint Conference on Artificial Intelligence (IJCAI) - Apr. 2021 - Present
Secretary, The Japan Society of Mechanical Engineers, Environmental Engineering Division, Thermoinformatics Working Group, Society - Apr. 2019 - Present
Steering committee, Special Interest Group on Multi‐Agent and Cooperative Computation (MACC), Japan Society for Software Science and Technology, Society - Jan. 2023 - Dec. 2023
委員, 第9回プライバシーワークショップ (PWS2023)実行委員会, Society - Nov. 2022 - Oct. 2023
担当委員, FIT 2023, Society - Jun. 2021 - Jun. 2023
Chair, Artificial Intelligence and Knowledge-Based Processing (AI), The Institute of Electronics, Information and Communication Engineers (IEICE), Society - Nov. 2021 - Oct. 2022
担当委員, FIT2022, Society - 01 Jun. 2018 - 31 May 2022
編集委員, 情報処理学会 論文誌ジャーナル/JIP編集委員会 - Nov. 2020 - Oct. 2021
プログラム委員, FIT2021, Society - 01 Apr. 2019 - 31 Mar. 2021
副委員長, 電子情報通信学会 人工知能と知識処理 研究専門委員会 - 2021 - 2021
Program Committee, International Joint Conference on Artificial Intelligence (IJCAI) - Nov. 2019 - Oct. 2020
担当委員, FIT2020, Society - 01 Apr. 2018 - 31 Mar. 2020
幹事, 情報処理学会 知能システム研究運営委員会 - Nov. 2018 - Oct. 2019
担当委員, FIT2019, Society - 01 Apr. 2017 - 31 Mar. 2019
幹事, 電子情報通信学会 人工知能と知識処理 研究専門委員会 - Apr. 2017 - Mar. 2019
Chair, Steering committee, Special Interest Group on Multi‐Agent and Cooperative Computation (MACC), Japan Society for Software Science and Technology, Society - 05 Nov. 2015 - 11 Mar. 2016
委員長, 経済産業省「先端課題に対応したベンチャー事業化支援等事業（ITベンチャー等によるイノベーション促進のための人材育成・確保モデル事業）」における「政府等におけるIT等調達等のスキル等に関する研究会」
Research Activity Information
Award
- Mar. 2024
The Telecommunications Advabncement Foundation
Privacy-Preserving Collaborative Data Collection and Analysis With Many Missing Values
The Telecommunications Advabncement Foundation Award, Yuichi Sei;J. Andrew Onesimu;Hiroshi Okumura;Akihiko Ohsuga
Publisher - Mar. 2024
第16回データ工学と情報マネジメントに関するフォーラム（DEIM）
バドミントンの試合データを用いたショットの成功確率予測
学生プレゼンテーション賞, 美濃岡 知樹
Japan society - Mar. 2024
第16回データ工学と情報マネジメントに関するフォーラム（DEIM）
意外性のある意見がもたらすエコーチェンバー現象の分析
学生プレゼンテーション賞, 中川 啓
Japan society - Mar. 2024
第16回データ工学と情報マネジメントに関するフォーラム（DEIM）
インターネットスラングを考慮した大規模言語モデルを用いた感情分析手法の提案
学生プレゼンテーション賞, 関 優花
Japan society - Mar. 2024
第16回データ工学と情報マネジメントに関するフォーラム（DEIM）
ポジティブな単語を含んだ煽り・誹謗中傷目的のコメント検出方法の提案
学生プレゼンテーション賞, 佐藤 豪
Japan society - Nov. 2023
IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS), The outstanding contribution was selected based on the thorough evaluation of originality and significance by the IEEE IoTaIS Program Committee.
Federated Learning Algorithm Handling Missing Attributes
Best Paper Award, Keiichiro Oishi;Yuichi Sei;Yasuyuki Tahara;Akihiko Ohsuga
International society - Sep. 2023
JAWS2023, JAWS2023の発表58件の中から最優秀賞１件、優秀賞４件、奨励賞10件が選出されました。
Analysis of Conditional Image Generation Methods Using Color Palettes in Animal Personification Task
JAWS Encouragement Award, Jianglin Xu;Ryohei Orihara;Yuchi Sei;Yasuyuki Tahara;Akihiko Ohsuga
Japan society - Sep. 2023
JAWS2023, JAWS2023の発表58件の中から最優秀賞１件、優秀賞４件、奨励賞10件が選出されました。
Estimation of Unmasked Face Images Based on Voice and 3DMM
JAWS Outstanding Paper Award, Tetsumaru Akatsuka;Ryohei Orihara;Yuichi Sei;Yasuyuki Tahara;Akihiko Ohsuga
Japan society - May 2023
JP生きがい振興財団, 警察職員による警察の科学技術に関する優れた研究論文を表彰するものであり本年は2件が選出
類似語を利用した複合語型隠語の検出
警察研究論文奨励賞 最優秀賞, 羽田 拓朗
Publisher - May 2023
The Japan Society of Refrigerating and Air Conditioning Engineers, 日本冷凍空調学会は、大正１４年に日本冷凍協会として、冷凍・冷蔵に関連する学術技術の発展と普及とを目的として設立以来、わが国の冷凍分野における唯一の公益法人として１００年近い歴史を歩んでいる。この学会において昨年1年間で採択された学術論文の中で最も優秀なものに与えられる賞である。
Heat Exchanger Optimization Using Genetic Refrigerant Flow Path Generation Algorithm
Science Award, Niccolo GIANNETTI;John Carlo;S. GARCIA;Richard Jayson VARELA;Yuichi Sei;Koji Enoki;Jongsoo Jeong;Kiyoshi Saito
International academic award - Mar. 2023
第15回データ工学と情報マネジメントに関するフォーラム（DEIM）
深層強化学習を用いた文章の言い換えによる駄洒落生成モデルの検討
学生プレゼンテーション賞, 南智仁
Japan society - Jan. 2023
情報処理学会, 情報処理学会論文誌ジャーナル／JIP特選論文は、情報処理学会論文誌ジャーナルおよびJIPに掲載された論文のうち、より多くの研究者が参照すべき論文に対して与えられる名称です。
ドメイン認証を用いた送信者レピュテーションの構築手法とフィードバックループの提案
情報処理学会論文誌ジャーナル/JIP特選論文, 櫻庭秀次;依田みなみ;清雄一;田原康之;大須賀昭彦 - Dec. 2022
電子情報通信学会「人工知能と知識処理」研究会
l-多様性を満たすためのグルーピングとダミー追加を組み合わせたアルゴリズム
研究奨励賞, 大石慶一朗;清雄一;田原康之;大須賀昭彦 - Sep. 2022
SMASH2022 Summer Symposium
Detection of compound-type dark jargon using similar words
Encouragement Award, Takuro Hada;Yuichi Sei;Yasuyuki Tahara;Akihiko Ohsuga
Japan society - May 2022
Funai Foundation
Research on privacy-preservving IoT data collection and analysis framework
Funai Information Technology Award, Yuichi Sei - Feb. 2022
電子情報通信学会「人工知能と知識処理」研究会
深層学習による汎用性の高いピアノリダクション自動生成技術
研究奨励賞, 星雄輝;折原良平;清雄一;田原康之;大須賀昭彦
Japan society - Feb. 2022
日本ソフトウェア科学会 マルチエージェントと協調計算研究会、情報処理学会 知能システム研究会、人工知能学会 データ指向構成マイニングとシミュレーション研究会
人の存在確率を考慮した位置情報プライバシ保護手法の提案
SMASH22 Winter Symposium 準優秀賞, 石禾里帆;清雄一;田原康之;大須賀昭彦
Japan society - Sep. 2021
SMASH2021 Summer Symposium
ファインチューニングを利用した少量音声からの韻律転送の試み
SMASH2021 Summer Symposium 奨励賞, 徳島大河;折原良平;清雄一;田原康之;大須賀昭彦
Japan society - Sep. 2021
SMASH2021 Summer Symposium
深層学習による汎用性を考慮したピアノリダクションの自動生成
SMASH2021 Summer Symposium 奨励賞, 星雄輝;折原良平;清雄一;田原康之;大須賀昭彦
Japan society - Aug. 2021
Interlayer Augmentation in a Classification Task
IEEE iCCECE2021 Best Paper Award, Satoru Mizusawa and Yuichi Sei
International society - Feb. 2021
IPSJ, IEEE Computer Society
Outstanding Research on Privacy-Preserving Web/IoT Data Analysis
IPSJ/IEEE Computer Society Young Computer Researcher Award, Yuichi Sei
International academic award - Oct. 2019
Information Processing Society of Japan
Linked Dataを用いた俯瞰的な多肢選択式問題自動生成手法の提案
IPSJ Specially Selected Paper, Fumika Okuhara;Yuichi Sei;Yasuyuki Tahara;Akihiko Ohsuga
Official journal - Dec. 2018
IEEE Computer Society Tokyo/Japan Joint Chapter
IEEE Computer Society Japan Chapter Young Author Award, Yuichi Sei - Dec. 2018
電子情報通信学会「人工知能と知識処理」研究会
2次元迷路課題における進化的計算を利用したマルチタスク深層強化学習
研究奨励賞, 今井翔太;清雄一;田原康之;大須賀昭彦
Japan society - Nov. 2018
IEEE Signal Processing Society Tokyo Joint Chapter, IEEE Signal Processing Society Kansai Chapter, and IEEE Signal Processing Society Sendai Chapter
IEEE Signal Processing Society (SPS) Japan Young Author Best Paper Award, Yuichi Sei
Official journal - Mar. 2018
公益財団法人 電気通信普及財団
電気通信普及財団賞テレコムシステム技術賞（奨励賞）, Yuichi Sei;Akihiko Ohsuga
Publisher - Dec. 2017
Asia Pacific Society for Computing and Information Technology (APSCIT)
APSCIT Outstanding Research Achievement Award, Yuichi Sei
International academic award - Jun. 2017
Information Processing Society of Japan
An Efficient Algorithm for Encrypted Text Searching in Cloud Computing
IPSJ Outstanding Paper Award, Yuichi Sei;Takao Takenouchi;Akihiko Ohsuga
Official journal - Mar. 2017
Committee on Hydroscience and Hydraulic Engineering
Development of the real-time river stage prediction method using deep learning
Best Paper Award, Masayuki Hitokoto;Masaaki Sakuraba;Yuichi Sei
Official journal - Mar. 2017
土木学会水工学委員会
深層学習を用いた河川水位予測手法の開発
平成28年度水工学論文賞, 一言正之;櫻庭雅明;清雄一
Official journal - Sep. 2016
家庭におけるペット-ロボットインタラクション～ロボットの世話行動による犬の行動変化の調査～
合同エージェントワークショップ＆シンポジウム(JAWS)優良論文賞, 鈴木もとこ;清雄一;田原康之;大須賀昭彦
Japan society - Sep. 2016
オートエンコーダを利用した複数話者の声質変換
合同エージェントワークショップ＆シンポジウム(JAWS)優良論文賞, 関井祐介;折原良平;小島圭介;清雄一;田原康之;大須賀昭彦
Japan society - Sep. 2016
オンラインレビューサイトにおけるレビュー解析精度向上に向けた皮肉文判別
合同エージェントワークショップ＆シンポジウム(JAWS)奨励論文賞, 鈴木翔太;折原良平;清雄一;田原康之;大須賀昭彦
Japan society - Mar. 2016
Linked Open Data Challenge Consortium
放置自転車LOD
Linked Open Data チャレンジ Japan 2015 データセット部門 最優秀賞, 江上周作;川村隆浩;清雄一;田原康之;大須賀昭彦 - Oct. 2015
Information Processing Society of Japan
クラウド上の安全で高速なキーワード検索アルゴリズムの提案
IPSJ Specially Selected Paper, Yuichi Sei;Takao Takenouchi;Akihiko Ohsuga
Official journal - Nov. 2014
Joint International Semantic Technology Conference Best Poster Award, Ryohei Yoko;Takahiro Kawamura;Yuichi Sei;Yasuyuki Tahara;Akihiko Ohsuga
International society - Nov. 2013
Joint International Semantic Technology Conference Best Poster & Demo Award, Kazuhiro Tashiro;Takahiro Kawamura;Yuichi Sei;Hiroyuki Nakagawa;Yasuyuki Tahara;Akihiko Ohsuga
International society - Sep. 2013
誤差を含む属性値のための柔軟な匿名データ収集
情報科学技術フォーラムFIT奨励賞, 清雄一
Japan society - Sep. 2007
無線センサーネットワークにおけるFalse Eventの検知
マルチメディア、分散、協調とモバイル (DICOMO)最優秀論文賞, 清雄一;松崎和賢;本位田真一 - Oct. 2006
合同エージェントワークショップ＆シンポジウム(JAWS)学生奨励賞, 清雄一 - Sep. 2006
Ringed Bloom Filterによる分散ハッシュテーブルのトラフィック量削減
マルチメディア、分散、協調とモバイル (DICOMO)優秀論文賞, 清雄一;松崎和賢;本位田真一
Paper
- Prediction of Boiling Heat Transfer Coefficients with Uncertainty under Upward Flow Conditions using Deep Neural Networks and Gaussian Process Regression
Tomihiro Kinjo; Koji Enoki; Yuichi Sei; Hayato Nakano
Proceedings of 10th World Conference on Experimental Heat Transfer, Fluid Mechanics and Thermodynamics, Aug. 2024, Peer-reviwed, True
International conference proceedings - A Model-Based Approach for Designing and Validating ABAC Policies
Duc-Hieu Nguyen; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 22nd IEEE/ACIS International Conference on Software Engineering, Management and Applications (SERA), May 2024, Peer-reviwed, True
International conference proceedings - The Proposal of Countermeasures for DeepFake Voices on Social Media Considering Waveform and Text Embedding
Yuta Yanagi; Ryohei Orihara; Yasuyuki Tahara; Yuichi Sei; Tanel Alumäe; Akihiko Ohsuga
Annals of Emerging Technologies in Computing, 8, 2, 15-31, Apr. 2024, Peer-reviwed, True, with international co-author(s)
Scientific journal - Analysis of the Echo Chamber Caused by Unexpected Opinions
Akira Nakagawa; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 7th International Conference on Information and Computer Technologies, Mar. 2024, Peer-reviwed, True
International conference proceedings - Integrating Behavioral, Biometric, and Environmental Data for Health Insights
Yuichi Sei
Lead, Proceedings of International Conference on Health Informatics, Intelligent Systems and Networking Technologies, Mar. 2024, Peer-reviwed, Invited, True
International conference proceedings - Minimizing Noise in Location Privacy Protection Through Equipment Error Consideration
Riho Isawa; Yuicih Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Journal of Electrical and Computer Engineering Systems, 15, 3, 285-296, Mar. 2024, Peer-reviwed, True
Scientific journal - Analysis of Conditional Image Generation Methods Using Color Palettes in Animal Personification Task
Jianglin Xu; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 7th International Conference on Information and Computer Technologies, Mar. 2024, Peer-reviwed - Hollowed-Out Icon Colorization with Controllable Diffusion Model
Koki Miyauchi; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 7th International Conference on Information and Computer Technologies, Mar. 2024, Peer-reviwed - Proposal of a Cosmetic Product Recommendation Method with Review Text that is Predicted to be Write by Users
Natsumi Baba; Yuicih Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 16th International Conference on Agents and Artificial Intelligence (ICAART), Feb. 2024, Peer-reviwed
International conference proceedings - An Analysis of Knowledge Representation for Anime Recommendation using Graph Neural Networks
Yuki Saito; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 16th International Conference on Agents and Artificial Intelligence (ICAART), Feb. 2024, Peer-reviwed
International conference proceedings - Algorithm to Satisfy l-diversity by Combining Dummy Records and Grouping
Keiichiro Oishi; Yuichi Sei; Andrew J; Yasuyuki Tahara; Akihiko Ohsuga
Security and Privacy, Feb. 2024, Peer-reviwed, True, with international co-author(s)
Scientific journal - Image editing with pre-trained StyleGAN using StyleMap
So Honda; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IPSJ Journal, 65, 1, 97-111, Jan. 2024, Peer-reviwed
Scientific journal - Diverse Level Generation for Tile-based Video Game using Generative Adversarial Networks from Few Samples
Soichiro Takata; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IPSJ Journal, 65, 1, 69-82, Jan. 2024, Peer-reviwed
Scientific journal - Background Image Editing Method by GAN Inversion with Semantic Segmentation
Shuusuke Ishihata; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IPSJ Journal, 65, 1, 83-96, Jan. 2024, Peer-reviwed
Scientific journal - Proposal of Model for Modifying Positioning of Football by Genetic Algorithm
Yuya Jingushi; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IPSJ Journal, 65, 1, 23-33, Jan. 2024, Peer-reviwed
Scientific journal - A Scalable Middleware for IoT Vulnerability Detection
Minami Yoda; Shigeo Nakamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 26th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), "Studies in Computational Intelligence," Springer, Dec. 2023, Peer-reviwed, True
International conference proceedings - Data-driven OCL Invariant Patterns-based Process Model Exploration for Process Mining
Duc-Hieu Nguyen; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 26th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), "Studies in Computational Intelligence," Springer, Dec. 2023, Peer-reviwed, True
International conference proceedings - Circuitry optimization using genetic programming for the advancement of next generation refrigerants
Niccolo Giannetti; John Carlo; S. Garcia; Cheol-Hwan Kim; Yuichi Sei; Koji Enoki; Kiyoshi Saito
International Journal of Heat and Mass Transfer, Elsevier, 217, 124648, 1-15, Dec. 2023, Peer-reviwed, True, with international co-author(s), In this study, a new evolutionary method, which can handle the implementation of genetic operators with unrestrained number and locations of splitting and merging nodes for the optimization of heat exchanger circuitries, is developed. Accordingly, this technique expands the search space of previous optimization studies. To this end, a finned-tube heat exchanger simulator is structured around a bijective mathematical representation of a refrigerant circuitry (the tube–tube adjacency matrix), which is used in combination with traversing algorithms from graph theory to recognize infeasible circuitries and constrain the evolutionary search to coherent and feasible offspring. The performance of three refrigerants, namely R32, R410A, and R454C, commonly used in air-conditioning applications was assessed for the optimized circuitries of a 36-tube evaporator while converging to a given cooling capacity, degree of superheating, and heat source boundary conditions. At a given output capacity and air outlet temperature, larger coefficient-of-performance improvements (up to 9.99% with reference to a common serpentine configuration) were realized for zeotropic refrigerant mixtures, such as R454C, where appropriate matching of the temperature glide with the temperature variation of the air yielded the possibility of further reducing the required compression ratio under the corresponding operating conditions. Hence, it was demonstrated that low-GWP zeotropic mixtures with temperature glide can realize a performance comparable to that of R32 and higher than that of R410A by approaching the Lorenz cycle operation.
Scientific journal - Federated Learning Algorithm Handling Missing Attributes
Keiichiro Oishi; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 6th IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS), Nov. 2023, Peer-reviwed, True
International conference proceedings - A Middleware to Improve Analysis Coverage in IoT Vulnerability Detection
Minami Yoda; Shigeo Nakamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 6th IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS), Nov. 2023, Peer-reviwed, True
International conference proceedings - Estimation of Unmasked Face Images Based on Voice and 3DMM
Tetsumaru Akatsuka; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 36th Australasian Joint Conference on Artificial Intelligence (AJCAI), 239-251, Sep. 2023, Peer-reviwed, True
International conference proceedings - Evolutionary optimization of heat exchanger circuitries for the advancement of next-generation refrigerants
Niccolo Giannetti; Adriano Milazzo; John Carlo; S. Garcia; Richard Jayson Varela; Yuichi Sei; Koji Enoki; Kiyoshi Saito
Proceedings of 26th International Congress of Refrigeration, Aug. 2023, Peer-reviwed, True, with international co-author(s)
International conference proceedings - Developing REM Sleep Prediction Models Using Smart Home Sensor Data
Atsuya Tsuda; Kazutaka Matsuzaki; Yuichi Sei
Last, Proceedings of IEEE World Conference on Applied Intelligence and Computing (AIC), Jul. 2023, Peer-reviwed, True
International conference proceedings - Blockchain for healthcare systems: Architecture, security challenges, trends and future directions
Andrew J; Deva Priya Isravel; K. Martin Sagayam; Bharat Bhushan; Yuichi Sei; Jennifer Eunice
Corresponding, Journal of Network and Computer Applications, Elsevier BV, 215, 103633, 1-36, Jun. 2023, Peer-reviwed, with international co-author(s)
Scientific journal - Privacy-Preserving Collaborative Data Collection and Analysis with Many Missing Values
Yuichi Sei; J. Andrew Onesimu; Hiroshi Okumura; Akihiko Ohsuga
Lead, IEEE Transactions on Dependable and Secure Computing, 20, 3, 2158-2173, May 2023, Peer-reviwed, with international co-author(s)
Scientific journal, English - Data collection of biomedical data and sensing information in smart rooms
Yuichi Sei; Akihiko Ohsuga
Lead, Data in Brief, Elsevier BV, 47, 108922, 1-18, Apr. 2023, Peer-reviwed, True
Scientific journal - Sign2Pose: A Pose-Based Approach for Gloss Prediction Using a Transformer Model
Jennifer Eunice; Andrew J; Yuichi Sei; D. Jude Hemanth
Sensors, MDPI AG, 23, 5, 2853:1-2853:23, Mar. 2023, Peer-reviwed, True, with international co-author(s), Word-level sign language recognition (WSLR) is the backbone for continuous sign language recognition (CSLR) that infers glosses from sign videos. Finding the relevant gloss from the sign sequence and detecting explicit boundaries of the glosses from sign videos is a persistent challenge. In this paper, we propose a systematic approach for gloss prediction in WLSR using the Sign2Pose Gloss prediction transformer model. The primary goal of this work is to enhance WLSR’s gloss prediction accuracy with reduced time and computational overhead. The proposed approach uses hand-crafted features rather than automated feature extraction, which is computationally expensive and less accurate. A modified key frame extraction technique is proposed that uses histogram difference and Euclidean distance metrics to select and drop redundant frames. To enhance the model’s generalization ability, pose vector augmentation using perspective transformation along with joint angle rotation is performed. Further, for normalization, we employed YOLOv3 (You Only Look Once) to detect the signing space and track the hand gestures of the signers in the frames. The proposed model experiments on WLASL datasets achieved the top 1% recognition accuracy of 80.9% in WLASL100 and 64.21% in WLASL300. The performance of the proposed model surpasses state-of-the-art approaches. The integration of key frame extraction, augmentation, and pose estimation improved the performance of the proposed gloss prediction model by increasing the model’s precision in locating minor variations in their body posture. We observed that introducing YOLOv3 improved gloss prediction accuracy and helped prevent model overfitting. Overall, the proposed model showed 17% improved performance in the WLASL 100 dataset.
Scientific journal - A k-Anonymization Method for Social Network Data with Link Prediction
Risa Sugai; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 9th International Conference on Information Systems Security and Privacy, SCITEPRESS - Science and Technology Publications, 493-500, Feb. 2023, Peer-reviwed
International conference proceedings - Rumor Detection in Tweets Using Graph Convolutional Networks
Takumi Takei; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 3, 397-404, Feb. 2023, Peer-reviwed
International conference proceedings - GAN Inversion with Editable StyleMap
So Honda; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 3, 389-396, Feb. 2023, Peer-reviwed
International conference proceedings - Diverse Level Generation for Tile-Based Video Game using Generative Adversarial Networks from Few Samples
Soichiro Takata; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 3, 326-333, Feb. 2023, Peer-reviwed
International conference proceedings - Proposal of a Signal Control Method Using Deep Reinforcement Learning with Pedestrian Traffic Flow
Akimasa Murata; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 3, 319-325, Feb. 2023, Peer-reviwed
International conference proceedings - Background Image Editing with HyperStyle and Semantic Segmentation
Syuusuke Ishihata; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 3, 293-300, Feb. 2023, Peer-reviwed
International conference proceedings - Predicting Visual Importance of Mobile UI Using Semantic Segmentation
Ami Yamamoto; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 3, 260-266, Feb. 2023, Peer-reviwed
International conference proceedings - Generation of Facial Images Reflecting Speaker Attributes and Emotions Based on Voice Input
Kotaro Koseki; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 2, 99-105, Feb. 2023, Peer-reviwed
International conference proceedings - Detection of Compound-Type Dark Jargons Using Similar Words
Takuro Hada; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 15th International Conference on Agents and Artificial Intelligence, SCITEPRESS - Science and Technology Publications, 1, 427-437, Feb. 2023, Peer-reviwed
International conference proceedings - Sender Reputation Construction Method And Feedback Loop Using Sender Authentication
Shuji Sakuraba; Minami Yoda; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
情報堀学会論文誌, 情報処理学会, 64, 1, 13-23, Jan. 2023, Peer-reviwed, 迷惑メール対策を目的として，送信ドメイン認証技術を用いた送信者レピュテーションの構築手法とフィードバックループについて提案する．送信者レピュテーションは，メールの送信元情報を用いて受け取りを判断するための情報であるが，その中で受け取るべき正規のメール送信元を収集することは簡単ではない．本論文では，転送メールの送信サーバが正規のメール送信元と考え，転送メールをメール受信時に送信ドメイン認証の結果を用いて判断する手法を示す．この転送メールの判断手法には課題が残っており，メール転送時に送信元情報を書き換える転送メールについては判断できない．本論文では，送信ドメイン認証の結果を用いて，メール転送時に送信元情報を書き換える転送メールについても判断できる手法を追加した，送信者レピュテーションの構築手法について示す．また送信者レピュテーションは，正規のメール送信元から送信される迷惑メールに対応できないという課題がある．この課題を改善するためには，迷惑メールの受信側からメールの送信元へ通知を行うフィードバックループが有効である．本論文では，通知の信頼性を向上させるために送信ドメイン認証を利用する手法を示す．送信者レピュテーションの構築手法の有効性を評価するために，実際のメールサービスでの受信記録情報を用いて適用し，より多くの受け取るべきメールを判定できることを示した．これらの手法により，迷惑メール対策において必要なメールがより確実に届き，送信者レピュテーションを悪用する正規のメールサーバの不正利用を改善することが可能となる．
We propose a sender reputation construction method and feedback loop using sender authentication technology for the purpose of preventing unsolicited emails. Sender reputation is information for judging receipt by using the sender information of mail, but it is not easy to collect the legitimate mail sender to be received in it. In this paper, the sender of the forwarded email is considered to be the legitimate email sender. We show a method to judge forwarded mail by using the result of sender authentication when receiving mail. There are still issues with this method of determining forwarded mail, and it is not possible to judge forwarded mail that rewrites the sender information when forwarding mail. In this paper, we show a method of constructing sender reputation by adding a method that can judge forwarded mail that rewrites the sender information at the time of mail forwarding by using the result of sender authentication. In addition, sender reputation has the problem that it cannot handle unsolicited emails sent from legitimate email sources. In order to improve this problem, a feedback loop that notifies the sender of the email from the receiver of the junk email is effective. In this paper, we show a method that uses sender authentication to improve the reliability of notifications. In order to evaluate the effectiveness of the sender reputation construction method, we applied it using the reception record information of the actual mail service, and showed that more mails to be received can be determined. With these methods, it is possible to more reliably receive the emails necessary for anti-spam measures, and to improve the unauthorized use of legitimate mail servers that abuse sender reputation.
Scientific journal, Japanese - Resource Provisioning Techniques in Multi-Access Edge Computing Environments: Outlook, Expression, and Beyond
S. Durga; Esther Daniel; J. Andrew Onesimu; Yuichi Sei
Last, Mobile Information Systems, Hindawi Limited, 2022, 1-24, 19 Dec. 2022, Peer-reviwed, True, with international co-author(s), Mobile cloud computing promises a research foundation in information and communication technology (ICT). Multi-access edge computing is an intermediate solution that reduces latency by delivering cloud computing services close to IoT and mobile clients (MCs), hence addressing the performance issues of mobile cloud computing. However, the provisioning of resources is a significant and challenging process in mobile cloud-based environments as it organizes the heterogeneous sensing and processing capacities to provide the customers with an elastic pool of resources. Resource provisioning techniques must meet quality of service (QoS) considerations such as availability, responsiveness, and reliability to avoid service-level agreement (SLA) breaches. This investigation is essential because of the unpredictable change in service demands from diverse regions and the limits of MEC’s available computing resources. In this study, resource provisioning approaches for mobile cloud computing are thoroughly and comparatively studied and classified as taxonomies of previous research. The paper concludes with an insightful summary that gives recommendations for future enhancements.
Scientific journal - Inspection of The Classifying Performance of The Deepfake Voices by The Latest Text-to-Speech Models
Yuta Yanagi; Ryohei Orihara; Yasuyuki Tahara; Yuichi Sei; Tanel Alumäe; Akihoko Ohsuga
2nd Interdisciplinary Conference on Mechanics, Computers and Electrics (ICMCE), 330-335, Oct. 2022, Peer-reviwed
International conference proceedings, English - Machine Learning Model Generation With Copula-Based Synthetic Dataset for Local Differentially Private Numerical Data
Yuichi Sei; J. Andrew Onesimu; Akihiko Ohsuga
Lead, IEEE Access, Institute of Electrical and Electronics Engineers (IEEE), 10, 101656-101671, Oct. 2022, Peer-reviwed, True, with international co-author(s)
Scientific journal, English - 遺伝的冷媒流路生成アルゴリズムを用いた熱交換器の最適化に関する研究
ジャンネッティ ニコロ; ガルシア ジョン カルロ; ヴァレラ リチャード; ジェイソン; 清 雄一; 榎木 光治; 鄭 宗秀; 齋藤 潔
日本冷凍空調学会論文集, 日本冷凍空調学会, 39, 3, 223-239, Sep. 2022, Peer-reviwed
Japanese - Proposal of a Middleware to Support Development of IoT Firmware Analysis Tools
Minami Yoda; Shuji Sakuraba; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
14th International Joint Conference on Knowledge-Based Software Engineering (JCKBSE), Learning and Analytics in Intelligent Systems, Springer, 30, 3-14, Aug. 2022, Peer-reviwed
International conference proceedings, English - Prediction of Boiling Heat Transfer Coefficients for Mini-Channels
Yuichi Sei; Koji Enoki; Seiichi Yamaguchi; Kiyoshi Saito
Lead, Multiphase Science and Technology, 34, 2, 43-65, May 2022, Peer-reviwed, Artificial intelligence (AI) techniques have been widely used across many fields. However, few studies have focused on the use of AI techniques for predicting heat transfer coefficients regardless of single-phase or two-phase flows. The applicability of deep neural networks [ (DNNs), also known as deep learning], one of the most promising AI techniques, to horizontal-flow boiling heat transfer in mini-channels is being actively researched. The effect of surface tension in mini-channels is significant in comparison to that in conventional large tubes, and the heat transfer mechanism in the mini-channels is complicated. Thus, the accuracy of the prediction results based on existing studies is not satisfactory. Moreover, we cannot determine the uncertainty of the predicted heat transfer coefficients by using existing approaches. In this study, we propose a novel prediction mechanism, based on the combination of a DNN and Gaussian process regression, that can predict not only heat transfer coefficients with high accuracy but also the uncertainties of the predicted heat transfer coefficients. We refer to this new research field, which integrates thermal engineering and informatics, as thermoinformatics, and consider the scope of its future development.
Scientific journal, English - Step-by-Step Acquisition of Cooperative Behavior in Soccer Task
Takashi Abe; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Journal of Advances in Information Technology, ENGINEERING & TECHNOLOGY PUBLISHING, 13, 2, 147-154, Apr. 2022, Peer-reviwed, Invited, In this research, soccer task is investigated among the numerous tasks of deep reinforcement learning. The soccer task requires cooperative behavior. However, it is difficult for the agents to acquire the behavior, because a reward is sparsely given. Moreover, the behaviors of the allies and opponents must be considered by the agents. In addition, in the soccer task, if the agents attempt to acquire high-level cooperative behavior from low-level movements, such as ball kicking, a huge amount of time will be needed to learn a model. In this research, we conduct experiments in which reward shaping and curriculum learning are incorporated into deep reinforcement learning. This enables the agents to efficiently acquire cooperative behavior from low-level movements in a soccer task. The findings of this research indicate that reward shaping and curriculum learning with a designer's domain knowledge positively influence the agent's attempt to acquire cooperative behavior from low-level movements.
Scientific journal, English - Prosody Transfer from a Small Amount of Voice using Fine Tuning
Taiga Tokushima; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Advanced Research in Computing (ICARC), 37-42, Feb. 2022, Peer-reviwed
International conference proceedings, English - Versatile Automatic Piano Reduction Generation System by Deep Learning
Yuki Hoshi; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Advanced Research in Computing (ICARC), 66-71, Feb. 2022, Peer-reviwed
International conference proceedings, English - コーパス間の類似語の差異に着目したマイクロブログにおける隠語検出
羽田拓朗; 清雄一; 田原康之; 大須賀昭彦
電気学会論文誌C, 142, 2, 177-189, Feb. 2022, Peer-reviwed
Scientific journal, Japanese - Multiobjective Geometry Optimization of Microchannel Heat Exchanger Using Real-Coded Genetic Algorithm
John Carlo Solomon Garcia; Niccolo Giannetti; Yuichi Sei; Kiyoshi Saito; Mamoru Houfuku; Ryoichi Takafuji
Applied Thermal Engineering, Elsevier BV, 202, 117821, 1-13, Feb. 2022, Peer-reviwed
Scientific journal, English - キャンパスオントロジーに基づく異種データ間の相関検出
塚越雄登; 江上周作; 清雄一; 田原康之; 大須賀昭彦
電気学会論文誌C, 141, 11, 1222-1233, Nov. 2021, Peer-reviwed
Scientific journal, Japanese - Sender Reputation Construction Method using Sender Authentication
Shuji Sakuraba; Minami Yoda; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IEEE International Conference on Data Science and Computer Application (ICDSCA), 369-373, Oct. 2021, Peer-reviwed, Spam is not just annoying, it is a serious problem that causes security problems. Mail filters are effective and widely used as anti-spam measures. However, spam senders are also becoming more sophisticated in their content and transmission methods, and countermeasures are becoming more difficult. In addition, if the mail filter makes a false judgment, there is also the problem that the necessary mail will not be delivered. In this paper, we propose a method for constructing sender reputation using sender authentication technologies. The sender of the forwarded mail is used as a method to determine the legitimate mail server. To determine the sender of the forwarded mail, use the authentication result of SPF and DKIM authentication. In addition, we propose a method using DKIM's block list as a countermeasure against forwarded spam. We used these methods to validate the sender reputation using the mails we actually received.
International conference proceedings, English - An Intelligent License Plate Detection and Recognition Model Using Deep Neural Networks
J. Andrew; Onesimu; Robin D. Sebastian; Yuichi Sei; Lenny Christopher
Corresponding, Annals of Emerging Technologies in Computing (AETiC), 5, 4, 23-36, Oct. 2021, Peer-reviwed
Scientific journal, English - Interlayer Augmentation in a Classification Task
Satoru Mizusawa; Yuichi Sei
4th IEEE International Conference on Computing, Electronics & Communications Engineering (iCCECE), 59-64, Aug. 2021, Peer-reviwed
International conference proceedings, English - Improvement of Legitimate Mail Server Detection Method using Sender Authentication
Shuji Sakuraba; Minami Yoda; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
18th IEEE/ACIS International Conference on Software Engineering, Management and Applications (SERA), IEEE, 10-14, Jun. 2021, Peer-reviwed
International conference proceedings - Sender Reputation Construction Method Using Sender Authentication Technologies
Shuji Sakuraba; Minami Yoda; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
情報処理学会論文誌, [出版社不明], 62, 5, 1173-1183, May 2021, Peer-reviwed, 迷惑メール対策には，メール内容から迷惑メールを判定する手法と送信者情報を用いる手法があげられる．送信者情報の送信元IPアドレスや送信者のドメイン名から受け取るべきメールかを判断できれば，判定のための処理負荷の高いメール内容によるメールフィルタの処理を軽減させることができる．本論文では，送信ドメイン認証技術を利用することで転送メールの送信元を特定し，メール転送元が受け取るべき送信元であることを示し，メール転送元を含めた正規メール送信元を収集することで許可リストを構築する手法を提案する．この手法を含めた，送信者レピュテーションの構築手法を提案し，実際に受信したメールの記録を利用して送信者レピュテーションを構築し，適用することで送信者レピュテーションの構築手法の有効性を示す．
Anti-spam measures include methods for determining unsolicited mail from the mail content and methods for using sender information. If it can be determined from the sender's IP address of sender information and the sender's domain name whether the email should be received, it is possible to reduce the processing of the email filter by the email content that has a high processing load for the determination. In this research, we identify senders of forwarded emails by using sender domain authentication technology, show that the senders should receive the emails, and collect the legitimate email senders including mail forwarders. We propose a method to build a permission list in. We propose a sender reputation construction method that includes this method, construct the sender reputation using the records of the actually received emails, and show the effectiveness of the sender reputation construction method.
Scientific journal, Japanese - Codeword Detection, Focusing on Differences in Similar Words Between Two Corpora of Microblogs
Takuro Hada; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Annals of Emerging Technologies in Computing, International Association for Educators and Researchers (IAER), 5, 2, 90-102, 01 Apr. 2021, Peer-reviwed, Recently, the use of microblogs in drug trafficking has surged and become a social problem. A common method applied by cyber patrols to repress crimes, such as drug trafficking, involves searching for crime-related keywords. However, criminals who post crime-inducing messages maximally exploit “codewords” rather than keywords, such as enjo kosai, marijuana, and methamphetamine, to camouflage their criminal intentions. Research suggests that these codewords change once they gain popularity; thus, effective codeword detection requires significant effort to keep track of the latest codewords. In this study, we focused on the appearance of codewords and those likely to be included in incriminating posts to detect codewords with a high likelihood of inclusion in incriminating posts. We proposed new methods for detecting codewords based on differences in word usage and conducted experiments on concealed-word detection to evaluate the effectiveness of the method. The results showed that the proposed method could detect concealed words other than those in the initial list and to a better degree than the baseline methods. These findings demonstrated the ability of the proposed method to rapidly and automatically detect codewords that change over time and blog posts that instigate crimes, thereby potentially reducing the burden of continuous codeword surveillance.
Scientific journal - Acquisition of Cooperative Behavior in a Soccer Task Using Reward Shaping
Takashi Abe; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
5th International Conference on Innovation in Artificial Intelligence (ICIAI), ACM, 145-150, Mar. 2021, Peer-reviwed
International conference proceedings - An Efficient Clustering-Based Anonymization Scheme for Privacy-Preserving Data Collection in IoT based Healthcare Services
J. Andrew Onesimu; J. Karthikeyan; Yuichi Sei
Peer-to-Peer Networking and Applications, 14, 3, 1629-1649, 2021, Peer-reviwed
Scientific journal - Detection of the Hardcoded Login Information from Socket and String Compare Symbols
Minami Yoda; Shuji Sakuraba; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Annals of Emerging Technologies in Computing, International Association for Educators and Researchers (IAER), 5, 1, 28-39, Jan. 2021, Peer-reviwed, Internet of Things (IoT) for smart homes enhances convenience; however, it also introduces the risk of the leakage of private data. TOP10 IoT of OWASP 2018 shows that the first vulnerability is ”Weak, easy to predict, or embedded passwords.” This problem poses a risk because a user can not fix, change, or detect a password if it is embedded in firmware because only the developer of the firmware can control an update. In this study, we propose a lightweight method to detect the hardcoded username and password in IoT devices using a static analysis called Socket Search and String Search to protect from first vulnerability from 2018 OWASP TOP 10 for the IoT device. The hardcoded login information can be obtained by comparing the user input with strcmp or strncmp. Previous studies analyzed the symbols of strcmp or strncmp to detect the hardcoded login information. However, those studies required a lot of time because of the usage of complicated algorithms such as symbolic execution. To develop a lightweight algorithm, we focus on a network function, such as the socket symbol in firmware, because the IoT device is compromised when it is invaded by someone via the Internet. We propose two methods to detect the hardcoded login information: string search and socket search. In string search, the algorithm finds a function that uses the strcmp or strncmp symbol. In socket search, the algorithm finds a function that is referenced by the socket symbol. In this experiment, we measured the ability of our proposed method by searching six firmware in the real world that has a backdoor. We ran three methods: string search, socket search, and whole search to compare the two methods. As a result, all methods found login information from five of six firmware and one unexpected password. Our method reduces the analysis time. The whole search generally takes 38 mins to complete, but our methods finish the search in 4-6 min.
Scientific journal - Stack performance improvement of stacked U-Net
Satoru Mizusawa; Yuichi Sei; Akihiko Ohsuga
9th IEEE Joint International Information Technology and Artificial Intelligence Conference (ITAIC), 2055-2060, Dec. 2020, Peer-reviwed
International conference proceedings - A Defense Method against Poisoning Attacks on IoT Machine Learning Using Poisonous Data
Tomoki Chiba; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
3rd IEEE International Conference on Artificial Intelligence and Knowledge Engineering (AIKE), IEEE, 84-91, Dec. 2020, Peer-reviwed
International conference proceedings - Ontology-Based Correlation Detection Among Heterogeneous Data Sets: A Case Study of University Campus Issues
Yuto Tsukagoshi; Shusaku Egami; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
3rd IEEE International Conference on Artificial Intelligence and Knowledge Engineering (AIKE), IEEE, 25-32, Dec. 2020, Peer-reviwed
International conference proceedings - Codewords Detection in Microblogs Focusing on Differences in Word Use Between Two Corpora
Takuro Hada; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
3rd IEEE International Conference on Computing, Electronics & Communications Engineering (iCCECE), IEEE, 103-108, Aug. 2020, Peer-reviwed, In recent years, drug trafficking using microblogs has risen and become a social problem. A common method of cyber patrols for cracking down on crimes, such as drug trafficking, involves searching for crime-related keywords. However, criminals who post crime-inducing messages make maximum use of "codewords" rather than keywords, such as enjo kosai, marijuana, and methamphetamine, to camouflage their criminal intentions. Research suggests that these codewords change once they become popular; therefore, searching for a specific word requires significant effort to keep track of the latest codewords. In this study, we focused on the appearance of codewords and those likely to be included in incriminating posts with aim to detect codewords with the high likelihood of inclusion in incriminating posts. We proposed new methods for detecting codewords based on differences in word usage and conducted experiments on concealed-word detection in order to evaluate method effectiveness. The results showed that the proposed method was capable of detecting concealed words other than those in the initial list and to better degree relative to baseline methods. These findings demonstrated the ability of the proposed method to rapidly and automatically detect codewords that change over time and blog posts that induce crimes, thereby potentially reducing the burden of continuous monitoring of codewords.
International conference proceedings, English - Detection of the hardcoded login information from socket symbols
Minami Yoda; Shuji Sakuraba; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
3rd IEEE International Conference on Computing, Electronics & Communications Engineering (iCCECE), IEEE, 33-38, Aug. 2020, Peer-reviwed, Internet of Things (IoT) for smart homes enhances the convenience of our life; however, it also introduces the risk of leakage of privacy data in the house. A user wants to protect their privacy data from leakage. However, the analysis of IoT devices requires technical knowledge; therefore, it is challenging for the users to detect any vulnerability by themselves. In this study, we propose a lightweight method to detect the hardcoded username and password in IoT devices using static analysis. This method can detect the 1st vulnerability from 2018 OWASP TOP 10 for the IoT device. The hardcoded login information can be obtained by comparing the user input with strcmp or strncmp. Thus, previous studies analyzed the symbols of strcmp or strncmp to detect the hardcoded login information. However, these studies require time because of the usage of complicated algorithms such as symbolic execution. To develop a lightweight algorithm, we focus on a network function, such as the socket symbol in firmware, because the IoT device is compromised when it is invaded by someone via the Internet. We propose two methods to detect the hardcoded login information, i.e., string search and socket search. In string searching, it finds a function that uses strcmp or strncmp symbol. In socket searching, it finds a function that is referenced by socket symbol. In the experiment, we measured the ability of our method by searching six firmware in the real world that has a backdoor. we ran three methods: string search, socket search, and whole search to compare two methods. As a result, all methods found login information from four of six firmware. Our method reduces an analysis time that when the whole search takes 38mins to complete, our methods finish 4-6min.
International conference proceedings, English - Proposal of Knowledge Graph and Completion Method for Solving Social Issues: An Attempt to Improve the Bicycle Parking Environment in the University
Yuto Tsukagoshi; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IEEJ Transactions on Electronics, Information and Systems, Institute of Electrical Engineers of Japan (IEE Japan), 140, 8, 905-915, Aug. 2020, Peer-reviwed
Scientific journal - A Method of Trajectory Anonymization with Adjustable Usefulness
Tomoki Chiba; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IEEJ Transactions on Electronics, Information and Systems, Institute of Electrical Engineers of Japan (IEE Japan), 140, 8, 956-963, Aug. 2020, Peer-reviwed
Scientific journal - Fake News Detection with Generated Comments for News Articles
Yuta Yanagi; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IEEE 24th International Conference on Intelligent Engineering Systems (INES), IEEE, 85-90, Jul. 2020, Peer-reviwed, Recently, fake news is shared via social networks and makes wrong rumors more diffusible. This problem is serious because the wrong rumor sometimes make social damage by deceived people. Fact-checking is a solution to measure the credibility of news articles. However the process usually takes a long time and it is hard to make it before their diffusion. Automatic detection of fake news is a popular researching topic. It is confirmed that considering not only articles but also social contexts(i.e. likes, retweets, replies, comments) supports to spot fake news correctly. However, the social contexts are naturally unavailable when an article comes out, making early fake news detection by means of the social context useless. We propose a fake news detector with the ability to generate fake social contexts, aiming to detect fake news in the early stage of its diffusion where few social contexts are available. The fake context generation is based on a fake news generator model. This model is trained to generate comments using a dataset which consists of news articles and their social contexts. In addition, we also trained a classify model. This used news articles, real-posted comments, and generated comments. To measure our detector' s effectiveness, we examined the performance of the generated comments for articles with real comments and generated ones by the classifying model. As a result, we conclude that considering a generated comment help detect more fake news than considering real comments only. It suggests that our proposed detector will be effective to spot fake news on social networks.
International conference proceedings, English - Differentially Private Mobile Crowd Sensing Considering Sensing Errors
Yuichi Sei; Akihiko Ohsuga
Lead, Sensors, MDPI AG, 20, 10, 2785-2785, May 2020, Peer-reviwed, An increasingly popular class of software known as participatory sensing, or mobile crowdsensing, is a means of collecting people’s surrounding information via mobile sensing devices. To avoid potential undesired side effects of this data analysis method, such as privacy violations, considerable research has been conducted over the last decade to develop participatory sensing that looks to preserve privacy while analyzing participants’ surrounding information. To protect privacy, each participant perturbs the sensed data in his or her device, then the perturbed data is reported to the data collector. The data collector estimates the true data distribution from the reported data. As long as the data contains no sensing errors, current methods can accurately evaluate the data distribution. However, there has so far been little analysis of data that contains sensing errors. A more precise analysis that maintains privacy levels can only be achieved when a variety of sensing errors are considered.
Scientific journal, English - Why Do Users Choose a Hotel over Others? Review Analysis Using Interpretation Method of Machine Learning Models
Takayuki Onogawa; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IEEE International Conference on Big Data Analytics (ICBDA), IEEE, 354-362, May 2020, Peer-reviwed, To date, existing research has attempted to extract user opinions relating to products and services by differentiating the products and services through their characteristics. However, it is difficult to obtain the characteristics of similar products and services in these studies. So-called competitive products and services are actually quite similar, though they might vary in their strengths and weaknesses across competitors. These advantages and disadvantages are essential for users when selecting products and services. On the other hand, in recent years, researchers have interpreted the output of machine learning models so that humans can understand the reason of the output. LIME and SP-LIME are typical approaches used in the literature. In this paper, we propose a new method using LIME and SP-LIME to compare the characteristics of the services of three competing hotels and investigate appropriate parameters for our method. We try to extract descriptive words for each competitor from review texts, taking a Japanese business hotel market and global game console market as examples.
International conference proceedings, English - Model smoothing using virtual adversarial training for speech emotion estimation using spontaneity
Toyoaki Kuwahara; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SCITEPRESS - Science and Technology Publications, 2, 587-594, Feb. 2020, Peer-reviwed
International conference proceedings, English - Hair Shading Style Transfer for Manga with cGAN
Masashi Aizawa; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SCITEPRESS - Science and Technology Publications, 2, 570-577, Feb. 2020, Peer-reviwed
International conference proceedings, English - Adaptation Plan Policy in Traffic Routing for Priority Vehicle
Krishna Priawan Hardinda; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Artificial Intelligence in Information and Communication (ICAIIC), IEEE, 113-118, Feb. 2020, Peer-reviwed
International conference proceedings, English - Hair Shading Style Transfer for Manga with cGAN
Masashi Aizawa; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SCITEPRESS - Science and Technology Publications, 1, 587-594, Feb. 2020, Peer-reviwed
International conference proceedings, English - Model Smoothing using Virtual Adversarial Training for Speech Emotion Estimation using Spontaneity
Toyoaki Kuwahara; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SCITEPRESS - Science and Technology Publications, 1, 570-577, Feb. 2020, Peer-reviwed
International conference proceedings, English - Generating Cooking Recipes from Cooking Videos Using Deep Learning Considering Previous Process with Video Encoding
Tatsuki Fujii; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the 3rd International Conference on Applications of Intelligent Systems, ACM, 21, 1-5, Jan. 2020, Peer-reviwed
International conference proceedings, English - Knowledge Graph of University Campus Issues and Application of Completion Methods
Yuto Tsukagoshi; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of 21st ACM International Conference on Information Integration and Web-based Applications & Services (iiWAS2019), ACM, 304-312, Dec. 2019, Peer-reviwed, (accepted)
International conference proceedings, English - 水位推定誤差の確率分布に基づく河川水位観測データのリアルタイム異常検知
一言正之; 川越典子; 橋田創; 清雄一; 房前和朋
土木学会論文集B1(水工学), 75, 2, 193-198, Nov. 2019, Peer-reviwed
Scientific journal, Japanese - Linked Dataを用いた俯瞰的な多肢選択式問題自動生成手法の提案
奥原史佳; 清雄一; 田原康之; 大須賀昭彦
情報処理学会論文誌, 60, 10, 1738-1756, Oct. 2019, Peer-reviwed
Scientific journal, Japanese - Self-Adaptation for Heterogeneous Client-Server Online Games
Satoru Yamagata; Hiroyuki Nakagawa; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Studies in Computational Intelligence (outstanding papers at 18th IEEE/ACIS International Conference on Computer and Information Science), Springer, 65-79, Aug. 2019, Peer-reviwed
International conference proceedings, English - Mis.Config: Finding Misreferred Configuration Bugs In Web Application Using Thin Slicing
Minami Yoda; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Studies in Computational Intelligence (outstanding papers at 18th IEEE/ACIS International Conference on Computer and Information Science), Springer, 47-64, Aug. 2019, Peer-reviwed
International conference proceedings, English - Do You Like Sclera? Sclera-region Detection and Colorization for Anime Character Line Drawings
Masashi Aizawa; Yuichi Sei; Yasuyuki Tahara; Ryohei Orihara; Akihiko Ohsuga
International Journal of Networked and Distributed Computing (IJNDC), 7, 3, 113-120, Aug. 2019, Peer-reviwed
Scientific journal, English - “Never fry carrots without chopping” Generating Cooking Recipes from Cooking Videos Using Deep Learning Considering Previous Process
Tatsuki Fujii; Yuichi Sei; Yasuyuki Tahara; Ryohei Orihara; Akihiko Ohsuga
International Journal of Networked and Distributed Computing (IJNDC), 7, 3, 107-112, Aug. 2019, Peer-reviwed
Scientific journal, English - Multi-task Deep Reinforcement Learning with Evolutionary Algorithm and Policy Gradients Method in 3D Control Tasks
Shota Imai; Yuichi Sei; Yasuyuki Tahara Yasuyuki; Ryohei Orihara; Akihiko Ohsuga
Studies in Computational Intelligence (outstanding papers at 4th IEEE/ACIS International Conference on Big Data, Cloud Computing, Data Science & Engineering), Springer, 19-32, Jul. 2019, Peer-reviwed
International conference proceedings, English - Trajectory Anonymization: Balancing Usefulness about Position Information and Timestamp
Tomoki Chiba; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IFIP International Conference on New Technologies, Mobility & Security (NTMS), IEEE, 1-6, Jun. 2019, Peer-reviwed
International conference proceedings, English - "Never fry carrots without cutting" Cooking Recipe Generation from Videos Using Deep Learning Considering Previous Process
Tatsuki Fujii; Yuichi Sei; Yasuyuki Tahara; Ryohei Orihara; Akihiko Ohsuga
4th IEEE/ACIS International Conference on Big Data, Cloud Computing, Data Science & Engineering, IEEE, 124-129, May 2019, Peer-reviwed
International conference proceedings, English - Do You Like the Sclera?: Sclera-Region Detection in Line Drawings for Automated Colorization
Masashi Aizawa; Yuichi Sei; Yasuyuki Tahara; Ryohei Orihara; Akihiko Ohsuga
4th IEEE/ACIS International Conference on Big Data, Cloud Computing, Data Science & Engineering, IEEE, 118-123, May 2019, Peer-reviwed
International conference proceedings, English - Model smoothing using virtual adversarial training for speech emotion estimation
Toyoaki Kuwahara; Yuichi Sei; Yasuyuki Tahara; Ryohei Orihara; Akihiko Ohsuga
4th IEEE/ACIS International Conference on Big Data, Cloud Computing, Data Science & Engineering, SCITEPRESS - Science and Technology Publications, 60-64, May 2019, Peer-reviwed
International conference proceedings, English - New Indicator for Centrality Measurements in Passing-network Analysis of Soccer
Masatoshi Kanbata; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 2, 616-623, Feb. 2019, Peer-reviwed
International conference proceedings, English - Transforming the Emotion in Speech using a Generative Adversarial Network
Kenji Yasuda; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 2, 427-434, Feb. 2019, Peer-reviwed
International conference proceedings, English - Generation of Multiple Choice Questions Including Panoramic Information using Linked Data
Fumika Okuhara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 1, 110-120, Feb. 2019, Peer-reviwed
International conference proceedings, English - Multi-task Deep Reinforcement Learning with Evolutionary Algorithm and Policy Gradients Method in 3D Control Tasks.
Shota Imai; Yuichi Sei; Yasuyuki Tahara; Ryohei Orihara; Akihiko Ohsuga
2019 IEEE International Conference on Big Data(BCD), IEEE, 100-105, 2019
International conference proceedings - サッカーPK戦におけるゲーム理論上の最適戦略とプロの戦略との差異に関する考察
小泉昂也; 折原良平; 清雄一; 田原康之; 大須賀昭彦
電子情報通信学会論文誌, J101-D, 9, 1363-1371, Sep. 2018, Peer-reviwed
Scientific journal, Japanese - パーチェス法を用いたエージェントシミュレーションによる金融機関の合併に関するシステミックリスクへの影響分析
加藤秀紀; 清雄一; 田原康之; 大須賀昭彦
電子情報通信学会論文誌, J101-D, 9, 1343-1353, Sep. 2018, Peer-reviwed
Scientific journal, Japanese - Anonymization and Analysis of Horizontally and Vertically Divided User Profile Databases with Multiple Sensitive Attributes
Yuki Ina; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
13th IEEE International Conference on Service Operations and Logistics, and Informatics (SOLI), IEEE, 262-267, Aug. 2018, Peer-reviwed
International conference proceedings, English - Text Classification and Transfer Learning based on Character-level Deep Convolutional Neural Networks
Minato Sato; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Agents and Artificial Intelligence (Revised selected papers from ICAART 2017), Springer, 60-81, Jun. 2018, Peer-reviwed, Invited
International conference proceedings, English - An Optimizing Placement of Passing Places in Mountainous Areas with Evolutionary Computing
Kazuhiro Amano; Munehiro Maeda; Yasuhiro Nakamura; Yuichi Sei; Akihiko Ohsuga
17th International Conference on Computing in Civil and Building Engineering (ICCCBE), 393-400, Jun. 2018, Peer-reviwed
International conference proceedings, English - 1.5車線的道路における待避区間の最適配置に向けた遺伝的アルゴリズム及び多目的最適化の適用
天野和洋; 前田宗宏; 中村泰広; 清雄一; 大須賀昭彦
土木学会論文集F3（土木情報学）, 73, 2, 109-117, Mar. 2018, Peer-reviwed
Scientific journal, Japanese - Proposal of l-diversity algorithm considering distance between sensitive attribute values
Keiichiro Oishi; Yasuyuki Tahara; Yuichi Sei; Akihiko Ohsuga
2017 IEEE Symposium Series on Computational Intelligence, SSCI 2017 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2018-, 1-8, 02 Feb. 2018, Peer-reviwed, Consideration of privacy is crucial when sharing a database that contains personal information with other organizations. Many organizations have utilized personal information while realizing the importance of personal privacy protection by anonymizing personal information according to existing indicators, such as k-anonymity. A database with personal information is defined as satisfying l-diversity when a specific record group that has the same combination of quasi-identifiers (QIDs) holds at least / kinds of sensitive attribute value. By satisfying l-diversity, the identification probability of the individual's sensitive attribute value becomes less than 1/l, and it can be said that privacy is protected. The l-diversity has been widely studied in the area of privacy-preserving data mining. However, if a database containing certain personal information holds similar sensitive attribute values, there is a possibility that de facto diversity is not satisfied, even if anonymization is performed to satisfy l-diversity. In this research, we propose (l, d)-semantic diversity that is able to consider more actual diversity to solve the problem of not being able to satisfy de facto diversity with the existing indicator. The (l, d)-semantic diversity considers the similarity of sensitive attribute values by adding distances, d, defined using categorization. We also propose an anonymization algorithm and analysis algorithm suitable for the proposal indicator, and we conduct evaluation experiments.
International conference proceedings, English - Factors Affecting Accuracy in Image Translation based on Generative Adversarial Network
Fumiya Yamashita; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 2, 446-453, 16 Jan. 2018, Peer-reviwed
International conference proceedings, English - Do Professional Football Players Follow the Optimal Strategies in Penalty Shootout?
Takaya Koizumi; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 2, 454-461, 16 Jan. 2018, Peer-reviwed
International conference proceedings, English - Agent-Based Simulation Model Embedded Accounting’s Purchase Method; Analysis on the Systemic Risk of Mergers and Acquisitions between Financial Institutions
Hidenori Kato; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 1, 168-175, 16 Jan. 2018, Peer-reviwed
International conference proceedings, English - Location Anonymization With Considering Errors and Existence Probability
Yuichi Sei; Akihiko Ohsuga
Lead, IEEE Transactions on Systems, Man, and Cybernetics: Systems, IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 47, 12, 3207-3218, Dec. 2017, Peer-reviwed, Mobile devices that can sense their location using GPS or Wi-Fi have become extremely popular. However, many users hesitate to provide their accurate location information to unreliable third parties if it means that their identities or sensitive attribute values will be disclosed by doing so. Many approaches for anonymization, such as k-anonymity, have been proposed to tackle this issue. Existing studies for k-anonymity usually anonymize each user's location so that the anonymized area contains k or more users. Existing studies, however, do not consider location errors and the probability that each user actually exists at the anonymized area. As a result, a specific user might be identified by untrusted third parties. We propose novel privacy and utility metrics that can treat the location and an efficient algorithm to anonymize the information associated with users' locations. This is the first work that anonymizes location while considering location errors and the probability that each user is actually present at the anonymized area. By means of simulations, we have proven that our proposed method can reduce the risk of the user's attributes being identified while maintaining the utility of the anonymized data.
Scientific journal, English - 人工知能の深層学習による円形微細流路内水平流の沸騰熱伝達の予測
榎木光治; 清雄一; 大川富雄; 齋藤潔
混相流, THE JAPANESE SOCIETY FOR MULTIPHASE FLOW, 31, 4, 412-421, Dec. 2017, Peer-reviwed, The applications of Artificial Intelligence ie AI show diversity in any fields. On the other hand, research of the predicting heat transfer regardless of single-phase or two-phase flow is still untouched. Therefore, we have confirmed usefulness using AI’s deep learning function on horizontal flow boiling heat transfer in flowing mini-channel that is actively researched. The effect of the surface tension in the mini-channel is large compared with conventional large tubes, and then the heat transfer mechanism is very complicated. For this reason, the numerical correlations of many existing researchers the prediction result is not good. However, the mechanistic correlation based on the visualization experiment, which the authors' research group published several years ago has very high precision. Therefore, in this research paper, we confirmed the effectiveness of using deep learning for predicting of the boiling heat transfer in mini-channel while comparing our correlation.
Scientific journal, Japanese - Development of the Real-Time River Stage Prediction Method Using Deep Learning
Masayuki Hitokoto; Masaaki Sakuraba; Yuichi Sei
Journal of JSCE, Division B: Hydraulic, Coastal and Environmental Engineering (Invited), 5, 11, 422-429, Dec. 2017, Invited
Scientific journal, English - 家庭におけるペット-ロボットインタラクション～ロボットのふるまいに対する犬の行動調査～
鈴木もとこ; 清雄一; 田原康之; 大須賀昭彦
情報処理学会論文誌, 58, 11, 1799-1807, Nov. 2017, Peer-reviwed, 今後ロボットが家庭に普及するために，人間とペットとロボットの3者が良い関係を築くことが大切である．我々は人とのインタラクションを目的に作られたロボットがペットと家庭で共生するために，ロボットにペットが好む行動をさせ，ペットがロボットをより好むようにすることを目標とする．本研究では，犬の世話行動をするロボットに注目した．世話行動をするケアロボットと世話行動をしないノンケアロボットの2台を用意し，ロボットが世話行動を行った後に犬がどちらのロボットをより好むか調査した．具体的には，ロボットが行う犬の世話行動は，飼い主へのアンケート結果をもとに餌やりとボール遊びの2種類とした．結果，犬は餌やりの世話行動をするケアロボットをノンケアロボットよりも有意に好むことが分かった．一方でボール遊びの世話行動については，ロボットとボール遊びをする犬としない犬の2群に分けられ，ボール遊びをする犬はボール遊びの世話行動をするケアロボットをノンケアロボットより有意に好むことが明らかになった．この知見は今後，家庭におけるロボットのペットに対する関わり方の指針となることが期待される．It is important for human beings, pets and robots to establish a good relationship in order for robots become popular to homes in the future. We are supposed to coexist between humans and robots that were made for human interaction at home. Then, let the robot take actions preferred by the pet and aim to make the robot more preferably by the pet. In this research, we focused on robots that take care of dogs. We used two robots that take care actions robot and non-care action robot, and surveyed which robot the dog likes more preferably after the robot takes care action. Specifically, the behavior of robot was two types of feeding and ball play based on the questionnaire result of the owner. As a result, it became clear that the dog significantly prefers the care robot which takes care of bait feeding than the non-care robot. On the other hand, as for the behavior of taking care of the ball play, it is divided into two groups: a dog which play ball with a robot, a dog which does not play ball with a robot. It became clear that dogs playing balls significantly prefer care robots to take care of ball playing than non-care robots. It is expected that this finding will be a guide for how robots are involved in pets at home.
Scientific journal, Japanese - Privacy-preserving Chi-squared testing for genome SNP databases
Yuichi Sei; Akihiko Ohsuga
Lead, 39th International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Institute of Electrical and Electronics Engineers Inc., 3884-3889, 13 Sep. 2017, Peer-reviwed, In recent years, the importance of privacy protection in genome-wide association studies (GWAS) has been increasing. GWAS focuses on identifying single-nucleotide polymorphisms (SNPs) associated with certain diseases such as cancer and diabetes, and Chi-squared testing can be used for this. However, recent studies reported that publishing the p-value or the corresponding chi-squared value of analyzed SNPs can cause privacy leakage. Several studies have been proposed for the anonymization of the chi-squared value with differential privacy, which is a de facto privacy metric in the cryptographic community. However, they can be applied to only small contingency tables
otherwise, they lose a lot of useful information. We propose novel anonymization methods: Rand-Chi and RandChiDist, and these methods are experimentally evaluated using real data sets.
International conference proceedings, English - 1.5車線的道路整備における待避区間の最適配置に向けた評価手法の検討
天野和洋; 前田宗宏; 中村泰広; 清雄一; 大須賀昭彦
土木学会論文集D3（土木計画学）, 73, 2, 124-134, Jun. 2017, Peer-reviwed
Scientific journal, Japanese - Differential Private Data Collection and Analysis Based on Randomized Multiple Dummies for Untrusted Mobile Crowdsensing
Yuichi Sei; Akihiko Ohsuga
Lead, IEEE Transactions on Information Forensics and Security, IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 12, 4, 926-939, Apr. 2017, Peer-reviwed, Mobile crowdsensing, which collects environmental information from mobile phone users, is growing in popularity. These data can be used by companies for marketing surveys or decision making. However, collecting sensing data from other users may violate their privacy. Moreover, the data aggregator and/ or the participants of crowdsensing may be untrusted entities. Recent studies have proposed randomized response schemes for anonymized data collection. This kind of data collection can analyze the sensing data of users statistically without precise information about other users' sensing results. However, traditional randomized response schemes and their extensions require a large number of samples to achieve proper estimation. In this paper, we propose a new anonymized data-collection scheme that can estimate data distributions more accurately. Using simulations with synthetic and real datasets, we prove that our proposed method can reduce the mean squared error and the JS divergence by more than 85% as compared with other existing studies.
Scientific journal, English - Japanese Text Classification by Character-level Deep ConvNets and Transfer Learning
Minato Sato; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
9th International Conference on Agents and Artificial Intelligence (ICAART), SCITEPRESS, 175-184, Feb. 2017, Peer-reviwed, Temporal (one-dimensional) Convolutional Neural Network ( Temporal CNN, ConvNet) is an emergent technology for text understanding. The input for the ConvNets could be either a sequence of words or a sequence of characters. In the latter case there are no needs for natural language processing that depends on a language such as morphological analysis. Past studies showed that the character-level ConvNets worked well for news category classification and sentiment analysis / classification tasks in English and romanized Chinese text corpus. In this article we apply the character-level ConvNets to Japanese text understanding. We also attempt to reuse meaningful representations that are learned in the ConvNets from a large-scale dataset in the form of transfer learning, inspired by its success in the field of image recognition. As for the application to the news category classification and the sentiment analysis and classification tasks in Japanese text corpus, the ConvNets outperformed N-gram-based classifiers. In addition, our ConvNets transfer learning frameworks worked well for a task which is similar to one used for pre-training.
International conference proceedings, English - Fast Many-to-One Voice Conversion using Autoencoders
Yusuke Sekii; Ryohei Orihara; Keisuke Kojima; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
ICAART: PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON AGENTS AND ARTIFICIAL INTELLIGENCE, VOL 2, SCITEPRESS, 164-174, 2017, Peer-reviwed, Most of voice conversion (VC) methods were dealing with a one-to-one VC issue and there were few studies that tackled many-to-one / many-to-many cases. It is difficult to prepare the training data for an application with the methods because they require a lot of parallel data. Furthermore, the length of time required to convert a speech by Deep Neural Network (DNN) gets longer than pre-DNN methods because the DNN-based methods use complicated networks. In this study, we propose a VC method using autoencoders in order to reduce the amount of the training data and to shorten the converting time. In the method, higher-order features are extracted from acoustic features of source speakers by an autoencoder trained with source speakers' data. Then they are converted to higher-order features of a target speaker by DNN. The converted higher-order features are restored to the acoustic features by an autoencoder trained with data drawn from the target speaker. In the evaluation experiment, the proposed method outperforms the conventional VC methods that use Gaussian Mixture Models (GMM) and DNNs in both one-to-one conversion and many-to-one conversion with a small training set in terms of the conversion accuracy and the converting time.
International conference proceedings, English - Sarcasm Detection Method to Improve Review Analysis
Shota Suzuki; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
ICAART: PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON AGENTS AND ARTIFICIAL INTELLIGENCE, VOL 2, SCITEPRESS, 519-526, 2017, Peer-reviwed, Currently, classifying sarcastic sentences into positive and negative sentiments has been a difficult problem and an important task. The sarcastic sentences could indicate negative meaning by using positive expressions, or positive meaning by using negative expressions. Sarcasm is a special kind of sentiment that comprise of words which mean the opposite of what you really want to say, especially in order to insult or wit someone, to show irritation, or to be funny. Therefore, determining sarcasm is an important task in order to correctly classify the sentence. In this paper, we propose an approach to detect sarcasm. First, we apply dependency parsing to amazon review data. After that, we classify phrases in the sentence into the proposed phrase based on the sequence of part-of-speech as proposed by Bharti et al. After being classified into either one of the phrase types, it is determined whether each phrase is positive or negative. If the emotions of the situation phrases and the sentiment phrases are different, the sentence is determined to be a "sarcasm". Using the above method, the experimental result shows the effectiveness of our method as compared with the the existing research.
International conference proceedings, English - An Observation of Behavioral Changes of Indoor Dogs in Response to Caring Behavior by Humanoid Robots Can Dogs and Robots Be Companions?
Motoko Suzuki; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
ICAART: PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON AGENTS AND ARTIFICIAL INTELLIGENCE, VOL 2, SCITEPRESS, 481-488, 2017, Peer-reviwed, The aim of our research is to build good relationships between pets and robots at home. We aim to promote of positive interaction between pets and robots. Recently, robots have been become popular with the general populace. There is a lot of research in human-robot interaction. We pay attention to pets that live in houses with humans. It is required for pets to like robots for positive interactions between pets and robots to exist. In this paper, we examine that 1) a robot can take care of dog, and 2) dogs and robots can be companion by caring behavior of robots toward dogs. In our experiment, we used two robots. One of the robots takes care of a dog, while the other does not. We observed which robot the dog chooses to interact with and had seventeen dogs participate in this study. We performed this statistical test to judge whether the dogs treated the robots with any significant differences.
International conference proceedings, English - User participatory construction of open hazard data for preventing bicycle accidents
Ryohei Kozu; Takahiro Kawamura; Shusaku Egami; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer Verlag, 10675, 289-303, 2017, Peer-reviwed, Recently, bicycle-related accidents, e.g., collision accidents at intersection increase and account for approximately 20% of all traffic accidents in Japan
thus, it is regarded as one of the serious social problems. However, the Traffic Accident Occurrence Map released by the Japanese Metropolitan Police Department is currently based on accident information records, and thus there are a number of near-miss events, which are overlooked in the map but will be useful for preventing the possible accidents. Therefore, we detect locations with high possibility of bicycle accidents using user participatory sensing and offer them drivers and government officials as Open Hazard Data (OHD) to prevent future bicycle accident. This paper uses smartphone sensors to obtain data for acceleration, location, and handle rotation information. Then, by classifying those data with convolutional neural networks, it was confirmed that the locations, where sudden braking occurred can be detected with an accuracy of 80%. In addition, we defined an RDF model for OHD that is currently publicly available. In future, we plan to develop applications using OHD, e.g., notifying alerts when users are approaching locations where near-miss events have occurred.
International conference proceedings, English - Linked Data Collection and Analysis Platform of Audio Features
Yuri Uehara; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Workshop and Poster of the 6th Joint International Semantic Technology Conference (JIST), CEUR-WS.org, 78-81, Nov. 2016, Peer-reviwed
International conference proceedings, English - Development of the Real-time River Stage Prediction Method Using Deep Learning
Masayuki HITOKOTO; Masaaki SAKURABA; Yuichi SEI
The Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering), 72, 4, 187-192, Feb. 2016, Peer-reviwed
Scientific journal, Japanese - Estimation of Interpersonal Relationships in Movies
Yuta Ohwatari; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG, E99D, 1, 128-137, Jan. 2016, Peer-reviwed, In many movies, social conditions and awareness of the issues of the times are depicted in any form. Even if fantasy and science fiction are works far from reality, the character relationship does mirror the real world. Therefore, we try to understand social conditions of the real world by analyzing the movie. As a way to analyze the movies, we propose a method of estimating interpersonal relationships of the characters, using a machine learning technique called Markov Logic Network (MLN) from movie script databases on the Web. The MLN is a probabilistic logic network that can describe the relationships between characters, which are not necessarily satisfied on every line. In experiments, we confirmed that our proposed method can estimate favors between the characters in a movie with F-measure of 58.7%. Finally, by comparing the relationships with social indicators, we discussed the relevance of the movies to the real world.
Scientific journal, English - Iterative Improvement of Human Pose Classification Using Guide Ontology
Kazuhiro Tashiro; Takahiro Kawamura; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG, E99D, 1, 236-247, Jan. 2016, Peer-reviwed, The objective of this paper is to recognize and classify the poses of idols in still images on the web. The poses found in Japanese idol photos are often complicated and their classification is highly challenging. Although advances in computer vision research have made huge contributions to image recognition, it is not enough to estimate human poses accurately. We thus propose a method that refines result of human pose estimation by Pose Guide Ontology (PGO) and a set of energy functions. PGO, which we introduce in this paper, contains useful background knowledge, such as semantic hierarchies and constraints related to the positional relationship between body parts. Energy functions compute the right positions of body parts based on knowledge of the human body. Through experiments, we also refine PGO iteratively for further improvement of classification accuracy. We demonstrate pose classification into 8 classes on a dataset containing 400 idol images on the web. Result of experiments shows the efficiency of PGO and the energy functions; the F-measure of classification is 15% higher than the non-refined results. In addition to this, we confirm the validity of the energy functions.
Scientific journal, English - Privacy Preservation for Participatory Sensing Applications
Yuichi Sei; Akihiko Ohsuga
Lead, IEEE 30TH INTERNATIONAL CONFERENCE ON ADVANCED INFORMATION NETWORKING AND APPLICATIONS IEEE AINA 2016, IEEE, 653-660, 2016, Peer-reviwed, Participatory sensing, which collects environmental information from mobile phone users, is growing in popularity. The collected information can be used for national policy or decision-making for companies. However, sensing users may violate their privacy. Recent studies have proposed negative surveys which can analyze the attributes of users statistically without precise information about each user's information. The traditional negative surveys need a lot of samples for proper estimation. These days, several types of negative surveys are used that can estimate the distribution of user attributes with a high degree of accuracy. However, privacy levels of these methods are relatively low. Moreover, existing studies assume that the privacy levels of all users are the same. In this paper, we propose a new negative survey that can estimate data distributions with more precision and can be used in a situation where the privacy levels are different based on each user's demand. By simulations of a synthetic and a real data set, we prove that our proposed method can estimate more precisely than existing methods.
International conference proceedings, English - Schema Design of Illegally Parked Bicycles LOD
Shusaku Egami; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
PROCEEDINGS 2016 5TH IIAI INTERNATIONAL CONGRESS ON ADVANCED APPLIED INFORMATICS IIAI-AAI 2016, IEEE, 692-697, 2016, Peer-reviwed, Illegally parked bicycles are a social problem in Japan and other countries. Illegally parked bicycles obstruct vehicles, cause road accidents, encourage thefts, and disfigure streets. In order to solve the challenge posed by illegally parked bicycles, we realized that it is necessary to collect and republish the data as reusable format. Therefore, we collected the number of illegally parked bicycles, location information, time, and factors. Then, we integrated and republished these data as Linked Open Data (LOD) on the Web. In this paper, we described a schema design of illegally parked bicycles LOD and a methodology of designing LOD schema. Furthermore, we collected data from SNS and website of municipality, and built the LOD of 21,898 triples.
International conference proceedings, English - A solution to visualize open urban data for illegally parked bicycles
Shusaku Egami; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer Verlag, 9860, 1, 129-141, 2016, Peer-reviwed, The illegal parking of bicycles is becoming an urban problem in Japan and other countries. We believe the data publication of such urban problems on the Web as Open Data will contribute to solving the problems. However, Open Data sets available for the illegally parked bicycles are coarse and in various formats, and then it is difficult to develop information services using the data. In this study, we thus build an ecosystem that generates Open Urban Data in Link Data format by socially collecting the data, complementing the missing data, and then visualizing the data to facilitate and raise social awareness of the problem. In our experiment, 747 pieces of information on the illegally parked bicycles in Tokyo were collected, and then we estimated the unknown number of the illegally parked bicycles with 64.3% accuracy. Then, we published the data as the Open Data, and also a web application, which visualizes the distribution of the illegally parked bicycles on a map.
International conference proceedings, English - Sake selection support application for countryside tourism
Teruyuki Iijima; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer Verlag, 9860, 1, 19-30, 2016, Peer-reviwed, For the upcoming Tokyo Olympic Paralympic Games in 2020, the number of foreign tourists coming to Japan is expected to rise. However, there has been a problem with tourists becoming less likely to visit places outside of the urban areas. In order to solve this issue, a commitment has been made by the government to use “Sake Brewery Tour” to draw tourists to less populated areas. The purpose of this study is to find a way to encourage foreign interest to sake and sake brewers, and participant in “Sake Brewery Tours”. We developed an application for the foreign tourists who are not much interested in sake. The approach of the study involved the presentation of sake selection in connection with wines, which have surprising similarities to the sakes, and encourage the tourists access sake brewer sites. 20 test users used the application, and the average screen residence time was 55 (sec) including the sake brewer sites, which was longer than the application for comparison, which shows the sake information alone. Therefore, we confirmed that the users come to have an interest in sake and sake brewers by showing the surprising connections with wine.
International conference proceedings, English - Linked data collection and analysis platform for music information retrieval
Yuri Uehara; Takahiro Kawamura; Shusaku Egami; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer Verlag, 10055, 127-135, 2016, Peer-reviwed, There has been extensive research on music information retrieval (MIR), such as signal processing, pattern mining, and information retrieval. In such studies, audio features extracted from music are commonly used, but there is no open platform for data collection and analysis of audio features. Therefore, we build the platform for the data collection and analysis for MIR research. On the platform, we represent the music data with Linked Data, which are in a format suitable for computer processing, and also link data fragments to each other. By adopting the Linked Data, the music data will become easier to publish and share, and there is an advantage that complex music analysis will be facilitated. In this paper, we first investigate the frequency of the audio features used in previous studies on MIR for designing the Linked Data schema. Then, we build a platform, that automatically extracts the audio features and music metadata from YouTube URIs designated by users, and adds them to our Linked Data DB. Finally, the sample queries for music analysis and the current record of music registrations in the DB are presented.
International conference proceedings, English - Privacy-Preserving Publication of Deep Neural Networks
Yuichi Sei; Hiroshi Okumura; Akihiko Ohsuga
Lead, IEEE International Conference on Data Science and Systems (DSS), IEEE, 1418-1425, 2016, Peer-reviwed, An organization that has a lot of personal data can create a deep neural network (DNN), which predicts sensitive attribute values such as the salary and diseases of people based on other attribute values such as age and hobbies. Moreover, by putting this data on the Cloud and providing the functionality of the DNN to other organizations, they can obtain new knowledge and can subsequently create new services. However, because such DNNs are generated from sensitive attribute values, we cannot share them freely without the explicit consent of the persons whose data are used for the DNNs. On the other hand, in recent years, epsilon-differential privacy has emerged as the de facto privacy metric. Many researchers use epsilon-differential privacy for privacy-preserving data mining such as correlation analysis and association rule analysis. In this paper, we modify epsilon-differential privacy for machine learning, and we propose three approaches for creating privacy-preserved DNNs based on the modified epsilon-differential privacy. Our proposed approaches are experimentally evaluated using a real data set, and we show that our approaches can protect personal attribute values while maintaining the accuracy of the DNNs.
International conference proceedings, English - A Survey on Perception of Privacy Metrics
Yuichi Sei; Midori Inaba; Akihiko Ohsuga
Lead, IPSJ journal, 56, 12, 2230-2243, Dec. 2015, Peer-reviwed
Scientific journal, Japanese - Transformation of KAOS Goal Models to BPMN Models Using Refinement Patterns
Hiroki Horita; Kozo Honda; Hideaki Hirayama; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
Computer Software, Japan Society for Software Science and Technology, 32, 4, 141-160, Nov. 2015, Peer-reviwed, In the software development, modeling business process is important. In constructing business process model appropriately, stakeholder's requirements should be reflected in the model. Therefore, in this research, we propose transformation approach from goal models using refinement pattern to business process models. It is denoted that rules of transformation and algorithm. Using our approach supports constructing business process models by specifying stakeholder's requirements formally using refinement patterns. We evaluate the effectiveness of our approach through applying our approach for a number of cases and using model-checking techniques.
Scientific journal, Japanese - An efficient algorithm for encrypted text searching in Cloud Computing
Yuichi Sei; Takao Takenouchi; Akihiko Ohsuga
Lead, IPSJ Journal, 56, 10, 1977-1987, Oct. 2015, Peer-reviwed
Scientific journal, Japanese - (1l, ...,1q)-diversity for Anonymizing Sensitive Quasi-Identifiers
Yuichi Sei; Takao Takenouchi; Akihiko Ohsuga
Lead, IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom), IEEE, 596-603, Aug. 2015, Peer-reviwed, A lot of studies of privacy-preserving data mining have been proposed. Most of them assume that they can separate quasi-identifiers (QIDs) from sensitive attributes. For instance, they assume that address, job, and age are QIDs but not sensitive attributes, and that a disease name is a sensitive attribute but not a QID. However, all of these attributes can have features that are both sensitive attributes and QIDs depending on the persons in practice. In this paper, we refer to these attributes as sensitive QIDs, and we propose a novel privacy definition (l1,..., lq)diversity and a method that can treat sensitive QIDs. Our method is composed of two algorithms: an anonymization algorithm and a reconstruction algorithm. The anonymization algorithm, which is conducted by data holders, is simple but effective, whereas the reconstruction algorithm, which is conducted by data users, can be conducted according to each data user's objective. Our proposed method is experimentally evaluated using real datasets.
International conference proceedings, English - News Curation Service using Semantic Graph Matching
Ryohei Yokoo; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Proc. of SEMAPRO 2015, IARIA, 2015, 25-31, 19 Jul. 2015, Peer-reviwed
International conference proceedings, English - Malicious node detection in mobilewireless sensor networks
Yuichi Sei; Akihiko Ohsuga
Lead, Journal of Information Processing, Information Processing Society of Japan, 23, 4, 476-487, Jul. 2015, Peer-reviwed, A compromised node in wireless sensor networks can be used to create false messages by generating them on their own or by falsifying legitimate messages received from other nodes. Because compromised nodes that create false messages can waste a considerable amount of network resources, we should detect them as early as possible. Existing studies for detecting such nodes can only be used in situations where sensor nodes do not move. However, it is possible that nodes move because of wind or other factors in real situations. We improve existing studies for detecting compromised nodes in mobile wireless sensor networks. In the proposed method, an agent exists on each node and it appends its ID and a k-bit code to an event message and the sink detects a compromised node by a statistical method. Our method can be used in static and dynamic environments. Simulations we conducted prove the effectiveness of our method.
Scientific journal, English - An algorithm for privacy-preserving location data collection by probabilistic dummy generation
Yuichi Sei; Akihiko Ohsuga
Lead, IEEJ Transactions on Electronics, Information and Systems, Institute of Electrical Engineers of Japan, 135, 6, 660-670, 01 Jun. 2015, Peer-reviwed, Mobile devices, which can sense their locations by GPS or Wi-Fi, have become popular these days, and we can collect and analyze location information of many users to examine traffic flow, conduct marketing analysis, and so on. However, several users hesitate to provide their accurate location information. Therefore, researches which anonymize user's location information on their devices and send the anonymized information to the data collection server have been proposed. These researches can protect user's privacy and let the data collection server to estimate the distribution of users' locations by a statistical way. However, they need many users to help with the data collection. In our proposed method each user sends several dummy locations to the data collection server and the server can estimate the location distribution with high accuracy. By mathematical analysis and simulations, we prove our proposed method can reduce the estimated errors by approximately from 85% to 99%.
Scientific journal, Japanese - 意外性のあるレシピを推薦するエージェントの提案
池尻恭介; 清雄一; 中川博之; 田原康之; 大須賀昭彦
電子情報通信学会論文誌, J98-D, 6, 971-981, Jun. 2015, Peer-reviwed
Scientific journal, Japanese - 語句間の意味的リレーションに基づくキュレーションエージェント
横尾亮平; 川村隆浩; 清雄一; 田原康之; 大須賀昭彦
電子情報通信学会論文誌, J98-D, 6, 982-991, Jun. 2015, Peer-reviwed
Scientific journal, Japanese - An Algorithm for l-diversity based on Randomized Addition of Sensitive Values
Yuichi Sei; Akihiko Ohsuga
Lead, 情報処理学会論文誌, 56, 5, 1377-1387, May 2015, Peer-reviwed
Scientific journal, Japanese - Analysis of flaming and its applications in CGM
Yuki Iwasaki; Ryohei Orihara; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
Transactions of the Japanese Society for Artificial Intelligence, Japanese Society for Artificial Intelligence, 30, 1, 152-160, 06 Jan. 2015, Peer-reviwed, Nowadays, anybody can easily express their opinion publicly through Consumer Generated Media. Because of this, a phenomenon of flooding criticism on the Internet, called flaming, frequently occurs. Although there are strong demands for flaming management, a service to reduce damage caused by a flaming after one occurs, it is very difficult to properly do so in practice. We are trying to keep the flaming from happening. It is necessary to identify the situation and the remark which are likely to cause flaming for our goal. Concretely, we propose methods to identify a potential tweet which will be a likely candidate of a flaming on Twitter, considering public opinion among Twitter users. Among three categories of flamings, our main focus is Struggles between Conflicting Values (SBCV), which is defined as a remark that forces one’s own opinion about a topic on others. Forecasting of this type of flamings is potentially desired since most of its victims are celebrities, who need to care one’s own social images. We proceed with a working hypothesis: a SBCV is caused by a gap between the polarity of the remark and that of public opinion. First, we have visualized the process how a remark gets flamed when its content is far from public opinion, by means of our original parameter daily polarity (dp). Second, we have built a highly accurate flaming prediction model with decision tree learning, using cumulative dp as an attribute along with parameters available from Twitter APIs. The experimental result suggests that the hypothesis is correct.
Scientific journal, Japanese - CGMにおける炎上の分析とその応用
岩崎 祐貴; 折原 良平; 清 雄一; 中川 博之; 田原 康之; 大須賀 昭彦
人工知能学会論文誌, (株) オーム社, 30, 1, 1-9, Jan. 2015, Peer-reviwed
Scientific journal, Japanese - Activity Recognition for Dogs Using Off-the-Shelf Accelerometer
Tatsuya Kiyohara; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 100-110, Jan. 2015, Peer-reviwed
International conference proceedings, English - Estimation of Character Diagram from Open-Movie Databases for Cultural Understanding
Yuta Ohwatari; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
2015 IEEE 9TH INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC), IEEE, 208-215, 2015, Peer-reviwed, In many movies, cultures, social conditions, and awareness of the issues of the times are depicted in any form. Even if fantasy and SF are works far from reality, the stories do mirror the real world. Therefore, we assumed to be able to understand social conditions and cultures of the real world by analyzing the movie. As a way to analyze the film, we decided to estimate the interpersonal relationships between the characters in the movies. In this paper, we propose a method of estimating interpersonal relationships of the characters using Markov Logic Network from movie script databases on the Web. Markov Logic Network is a probabilistic logic network that can describe the relationships between characters, which are not necessarily satisfied on every occasion. In experiments, we confirmed that our proposed method can estimate favors between the characters in a movie with a precision of 64.2%. Finally, by comparing the estimated relationships with social indicators, we discussed the relevance of the movie to the real world.
International conference proceedings, English - Decision Making Strategy Based on Time Series Data of Voting Behavior
Shogo Higuchi; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
AI 2015: ADVANCES IN ARTIFICIAL INTELLIGENCE, SPRINGER-VERLAG BERLIN, 9457, 229-241, 2015, Peer-reviwed, In gambling such as horse racing, we are sometimes able to peep peculiar voting behavior by a punter with the advantageous information closely related to the results. The punter is often referred as an insider. In this study, our goal is to propose a reasonable investment strategy by peeping insiders' decision-making based on the time series odds data in horse racing events held by JRA. We have found the conditions that the rate of return is more than 642 % for races whose winner's prize money is 20 million yens or more. That suggests the possibility of Knowledge Peeping.
International conference proceedings, English - Towards the Elimination of the Miscommunication Between Users in Twitter
Tomoaki Ueda; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
AI 2015: ADVANCES IN ARTIFICIAL INTELLIGENCE, SPRINGER-VERLAG BERLIN, 9457, 589-595, 2015, Peer-reviwed, In recent years, a Twitter response from another user who does not share the intentions and expectations of the original poster may cause discomfort and stress, which is a social phenomenon known as SNS fatigue. For example, a user may receive answers that are different from her/his expectation after the user posts a question on the timeline. In the background of such responses there is a miscommunication between users. In order to resolve the problem, it is important to know what the original poster expected as responses to her/his tweet. In this paper, we propose a classification method of tweets according to the response that users expect, and experimentally evaluate it. As a result, we have shown that tweets which the poster does not expect any replies can be classified with 76.2 % of the average precision.
International conference proceedings, English - Activity Recognition for Dogs Based on Time-series Data Analysis
Tatsuya Kiyohara; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
AGENTS AND ARTIFICIAL INTELLIGENCE, ICAART 2015, SPRINGER INT PUBLISHING AG, 9494, 163-184, 2015, Peer-reviwed, Dogs are one of the most popular pets in the world, and more than 10 million dogs are bred annually in Japan now [4]. Recently, primitive commercial services have been started that record dogs' activities and report them to their owners. Although it is expected that an owner would like to know the dog's activity in greater detail, a method proposed in a previous study has failed to recognize some of the key actions. The demand for their identification is highlighted in responses to our questionnaire. In this paper, we show a method to recognize the actions of the dog by attaching only one off-the-shelf acceleration sensor to the neck of the dog. We apply DTW-D which is the state-of-the-art time series data search technique for activity recognition. Application of DTW-D to activity recognition of an animal is unprecedented according to our knowledge, and thus is the main contribution of this study. As a result, we were able to recognize eleven different activities with 75.1% classification F-measure. We also evaluate the method taking account of real-world use cases.
International conference proceedings, English - Locating Malicious Agents in Mobile Wireless Sensor Networks
Yuichi Sei; Akihiko Ohsuga
Lead, Principles and Practice of Multi-Agent Systems (PRIMA), SPRINGER-VERLAG BERLIN, 8861, 206-221, Dec. 2014, Peer-reviwed, A compromised node in wireless sensor networks can be used to create false messages by generating them on their own or by fabricating legitimate messages received from other nodes. Our goal is to locate the compromised nodes that create false messages. Existing studies can only be used in situations where sensor nodes do not move. However, it is possible that nodes move because of wind or other factors in real situations. We improve existing studies for detecting compromised nodes in mobile wireless sensor networks. In the proposed method, an agent exists on each node and it appends its ID and a k-bit code to an event message and the sink detects a compromised node by a statistical method. Our method can be used in static and dynamic environments. Simulations we conducted prove the effectiveness of our method.
International conference proceedings, English - 希少性と一般性に基づいた意外性のある食材の抽出
池尻恭介; 清雄一; 中川博之; 田原康之; 大須賀昭彦
日本ソフトウェア科学会コンピュータソフトウェア, 31, 3, 70-78, Sep. 2014, Peer-reviwed
Scientific journal, Japanese - Human Pose Guide Ontologyを用いた静止画像内人物のポーズ分類
田代和浩; 川村隆浩; 清雄一; 中川博之; 田原康之; 大須賀昭彦
日本ソフトウェア科学会コンピュータソフトウェア, 31, 3, 58-69, Sep. 2014, Peer-reviwed
Scientific journal, Japanese - 多次元属性のための匿名データ収集アルゴリズムの提案
清雄一; 大須賀昭彦
情報処理学会論文誌, 55, 9, 2120-2133, Sep. 2014, Peer-reviwed
Scientific journal, Japanese - Surprising Ingredient Extraction based on Rarity and Generality
Kyosuke Ikejiri; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
JSSST Journal Compupter Software, Jul. 2014, Peer-reviwed
Scientific journal, Japanese - 誤差を考慮した位置匿名化手法の提案
清雄一; 大須賀昭彦
Lead, 電子情報通信学会論文誌, 97-D, 5, 964-974, May 2014, Peer-reviwed
Scientific journal, Japanese - Randomized Responseを用いた柔軟な匿名データ収集
清雄一; 大須賀昭彦
Lead, 電子情報通信学会論文誌, 97-D, 5, 953-963, May 2014, Peer-reviwed
Scientific journal, Japanese - ユビキタスコンピューティングにおけるl-エントロピーを満たす匿名データ収集
清雄一; 大須賀昭彦
Lead, 電子情報通信学会論文誌, 97-D, 4, 793-806, Apr. 2014, Peer-reviwed
Scientific journal, Japanese - Surprising Recipe Extraction Based on Rarity and Generality of Ingredients
Kyosuke Ikejiri; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 428-436, Mar. 2014, Peer-reviwed
International conference proceedings, English - Identification of Flaming and Its Applications in CGM: Case Studies toward Ultimate Prevention
Yuki Iwasaki; Ryohei Orihara; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), SciTePress, 639-644, Mar. 2014, Peer-reviwed
International conference proceedings, English - Context-Aware Music Recommendation with Serendipity Using Semantic Relations
Mian Wang; Takahiro Kawamura; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
SEMANTIC TECHNOLOGY, SPRINGER-VERLAG BERLIN, 8388, 17-32, 2014, Peer-reviwed, A goal for the creation and improvement of music recommendation is to retrieve users' preferences and select the music adapting to the preferences. Although the existing researches achieved a certain degree of success and inspired future researches to get more progress, problem of the cold start recommendation and the limitation to the similar music have been pointed out. Hence we incorporate concept of serendipity using 'renso' alignments over Linked Data to satisfy the users' music playing needs. We first collect music-related data from Last. fm, Yahoo! Local, Twitter and LyricWiki, and then create the 'renso' relation on the Music Linked Data. Our system proposes a way of finding suitable but novel music according to the users' contexts. Finally, preliminary experiments confirm balance of accuracy and serendipity of the music recommendation.
International conference proceedings, English - Transformation approach from KAOS goal models to BPMN models using refinement patterns
Hiroki Horita; Kozo Honda; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
Proceedings of the ACM Symposium on Applied Computing, Association for Computing Machinery, 1023-1024, 2014, Peer-reviwed, It is important to make sure that software satisfies stakeholders' requirements. However, as software has been getting more large-scale and complicated in recent years, it has certainly become more difficult to satisfy requirements. Even if there are goal-oriented requirements analysis approaches as techniques to systematically satisfy stakeholders' requirements, it is still difficult to maintain consistency of goal models and other requirements artifacts. In this paper, we propose a transformation approach that transforms models of KAOS, a well-known goal modeling methodology, into preliminary BPMN models by using refinement pattern of KAOS in a systematic way. It can assure consistency between systematically defined user's requirements and their realization process. Copyright 2014 ACM.
International conference proceedings, English - Music recommender adapting implicit context using 'renso' relation among Linked Data
Mian Wang; Takahiro Kawamura; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
Journal of Information Processing, Information Processing Society of Japan, 22, 2, 279-288, 2014, Peer-reviwed, The existing music recommendation systems rely on user's contexts or content analysis to satisfy the users' music playing needs. They achieved a certain degree of success and inspired future researches to get more progress. However, a cold start problem and the limitation to the similar music have been pointed out. Therefore, this paper proposes a unique recommendation method using a 'renso' alignment among Linked Data, aiming to realize the music recommendation agent in smartphone. We first collect data from Last.fm, Yahoo! Local, Twitter and LyricWiki, and create a large scale of Linked Open Data (LOD), then create the 'renso' relation on the LOD and select the music according to the context. Finally, we confirmed an evaluation result demonstrating its accuracy and serendipity. © 2014 Information Processing Society of Japan.
Scientific journal, English - Refinement of Ontology-constrained Human Pose Classification
Kazuhiro Tashiro; Takahiro Kawamura; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
2014 IEEE INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC), IEEE, 60-67, 2014, Peer-reviwed, In this paper, we propose an image classification method that recognizes several poses of idol photographs. The proposed method takes unannotated idol photos as input, and classifies them according to their poses based on spatial layouts of the idol in the photos. Our method has two phases; the first one is to estimate the spatial layout of ten body parts (head, torso, upper and lower arms and legs) using Eichner's Stickman Pose Estimation. The second one is to classify the poses of the idols using Bayesian Network classifiers. In order to improve accuracy of the classification, we introduce Pose Guide Ontology (PGO). PGO contains useful background knowledge, such as semantic hierarchies and constraints related to the positional relationship between the body parts. The location information of body parts is amended by PGO. We also propose iterative procedures for making further refinements of PGO. Finally, we evaluated our method on a dataset consisting of 400 images in 8 poses, and the final results indicated that F-measure of the classification has become 15% higher than non-amended results.
International conference proceedings, English - Randomized Addition of Sensitive Attributes for l-diversity
Yuichi Sei; Akihiko Ohsuga
Lead, 2014 11TH INTERNATIONAL CONFERENCE ON SECURITY AND CRYPTOGRAPHY (SECRYPT), IEEE, 350-360, 2014, Peer-reviwed, When a data holder wants to share databases that contain personal attributes, individual privacy needs to be considered. Existing anonymization techniques, such as l-diversity, remove identifiers and generalize quasiidentifiers (QIDs) from the database to ensure that adversaries cannot specify each individual's sensitive attributes. Usually, the database is anonymized based on one-size-fits-all measures. Therefore, it is possible that several QIDs that a data user focuses on are all generalized, and the anonymized database has no value for the user. Moreover, if a database does not satisfy the eligibility requirement, we cannot anonymize it by existing methods. In this paper, we propose a new technique for l-diversity, which keeps QIDs unchanged and randomizes sensitive attributes of each individual so that data users can analyze it based on QIDs they focus on and does not require the eligibility requirement. Through mathematical analysis and simulations, we will prove that our proposed method for l-diversity can result in a better tradeoff between privacy and utility of the
International conference proceedings, English - A MAPE Loop Control Pattern for Heterogeneous Client/Server Online Games
Satoru Yamagata; Hiroyuki Nakagawa; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Software Engineering & Knowledge Engineering (SEKE), Knowledge Systems Institute Graduate School, 742-743, 2014, Peer-reviwed
International conference proceedings, English - Towards Software Evolution for Embedded Systems Based on MAPE Loop Encapsulation
Hiroyuki Nakagawa; Takumitsu Kudo; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
2014 IEEE EIGHTH INTERNATIONAL CONFERENCE ON SELF-ADAPTIVE AND SELF-ORGANIZING SYSTEMS (SASO), IEEE, 203-204, 2014, Peer-reviwed, Software evolution is an essential activity that adapts existing software to changes in requirements. Because of recent rapid requirements changes, systems are strongly required to evolve even if the target systems are embedded systems, whose implementation code is generally hard to be changed. This paper discusses the feasibility of applying self-adaptation mechanism for software evolution. We use the MAPE loop mechanism to evolve embedded systems without changing code inside the existing systems. This paper also reports preliminary results that we experimentally evolved a cleaning robot by following our approach. Our demonstrations show a part of additional behaviors as the results of software evolution that makes the cleaning robot possible to move obstacles. We also discuss the future directions of software evolution for embedded systems with the self-adaptive mechanism.
International conference proceedings, English - Estimation of Character Diagram from Open Movie Database using Markov Logic Network
Yuta Ohwatari; Takahiro Kawamura; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
4th Joint International Semantic Technology (JIST), CEUR-WS.org, 124-127, 2014, Peer-reviwed
International conference proceedings, English - メディア情報のLinked Data 化と活用事例の提案
川村隆浩; 越川兼地; 中川博之; 清雄一; 田原康之; 大須賀昭彦
電子情報通信学会論文誌, The Institute of Electronics, Information and Communication Engineers, J96-D, 12, 2987-2999, Dec. 2013, Peer-reviwed, 昨今,インターネットの普及などから様々な情報源(ソーシャルメディア・マスメディア)に容易にアクセスし,多様な意見・考え方に触れることが可能になった.同時に,ソーシャルメディアにおけるデマの拡散や,マスメディアにおける偏向報道・情報操作の疑いなど,ユーザ自身が情報の信頼性について自ら判断することが求められてきている.そこで我々は,一般ユーザがメディア情報を多角的な観点から比較することを支援するため,ユーザに代わってソーシャル,マス両メディアから特定の話題に関する情報を抽出,見える化し,特定の観点に基づく比較ポイントを提示するエージェントシステムを目指している.本論文では,Conditional Random Fieldsと事象抽出のためのヒューリスティクスを用いて,Twitter上のツイート,マスメディアのニュース記事等から13の属性情報をもつ事象情報を抽出し,それらをLinked Data化する手法を提案し,精度評価を行った.また,事例を通して多様性,希少性,偏在性,因果関係の四つの観点に沿って比較ポイントを抽出することで有用性を確認した.
Scientific journal, Japanese - Context-aware Music Recommendation with Serendipity Using Semantic Relations
Mian Wang; Takahiro Kawamura; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
Proc. The 3rd Joint International Semantic Technology Conference (JIST2013), JIST2013, 1-16, Nov. 2013, Peer-reviwed
International conference proceedings, English - Classification of Idol Photography Based on Pose Guide Ontology
Kazuhiro Tashiro; Takahiro Kawamura; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
Proc. Joint International Semantic Technology (JIST) conference (poster), PD4, Oct. 2013, Peer-reviwed
International conference proceedings, English - Towards semi-automatic identification of functional requirements in legal texts for public administration
Yutaka Yoshida; Kozo Honda; Yuichi Sei; Hiroyuki Nakagawa; Yasuyuki Tahara; Akihiko Ohsuga
Frontiers in Artificial Intelligence and Applications, 259, 175-184, 2013, Peer-reviwed, There is a need for the development of systems that are compliant with laws in public administration, because their administrative activities are based on laws. When new laws are made or existing laws are amended, however, civil servants need to develop or modify the systems in the short time before the laws are issued. Related work in requirements elicitation from the legal texts includes approaches using ontology but there are difficulties in building an ontology for practical use. In this paper we propose pre-defined templates with the expression of functional requirements to identify legal texts, including their functional requirements, and a support tool consisting of two functions, one for automatic summary creation from complicated legal texts and one for the suggestion of the legal texts, including their functional requirements. We have also applied this approach to Japanese laws and have evaluated its accuracy. Our research revealed that using this approach can identify functional requirements with high accuracy. © 2013 The authors and IOS Press.
International conference proceedings, English - False event detection for mobile sinks in wireless sensor networks
Yuichi Sei; Akihiko Ohsuga
Proceedings - 2013 European Intelligence and Security Informatics Conference, EISIC 2013, 52-59, 2013, Peer-reviwed, In large-scale sensor networks, adversaries may capture and compromise several of the sensors. Compromised nodes can be used by adversaries to generate many false messages which waste the batteries of sensor nodes and the bandwidth of the sensor network. Many works aim to detect a false event in-network even if many nodes are compromised. Certain existing methods can achieve this, but, they cannot be used in a situation where the location of the sink changes. We propose a new method that resiliently detects false messages, even when there are a large number of compromised nodes and that can handle situations where the location of the sink changes. By preloading a legitimate combination of keys (LCK) on sensor nodes before deployment, the nodes can detect false events created from false combinations of keys. Our mathematical analysis and the simulations we conducted prove the effectiveness of our method. © 2013 IEEE.
International conference proceedings, English - False Event Detection for Rare Events in Wireless Sensor Networks
Yuichi Sei; Akihiko Ohsuga
Proc. of 6th International Conference on Advanced Computer Theory and Engineering (ICACTE), Article No. 5 (8 pages), 2013, Peer-reviwed
International conference proceedings, English - Need Only One Bit: Light-weight Packet Marking for Detecting Compromised Nodes in WSNs
Yuichi Sei; Akihiko Ohsuga
Proc. of The Seventh International Conference on Emerging Security Information, Systems and Technologies (SECURWARE), 134-143, 2013, Peer-reviwed
International conference proceedings, English - 多数のノード取得攻撃に対応した無線センサーネットワークにおける複製ノードの分散検知
清 雄一; 本位田 真一
Lead, 電子情報通信学会論文誌, 92-B, 4, 689-699, Mar. 2009, Peer-reviwed
Scientific journal, Japanese - 多数のノード取得攻撃に対応した無線センサーネットワークにおける不正イベントの検知
清雄一; 本位田真一
Lead, 電子情報通信学会論文誌, 92-B, 4, 678-688, Mar. 2009, Peer-reviwed
Scientific journal, Japanese - 無線センサーネットワークにおける不正メッセージ作成元ノードの検知
清雄一; 本位田真一
Lead, 情報処理学会論文誌, 情報処理学会, 50, 2, 787-797, Feb. 2009, Peer-reviwed
Scientific journal, Japanese - 分散配列:効率的な論理配列を実現するP2Pデータ構造
福地 大輔; クリスチャンソッメル; 清 雄一; 本位田真一
情報処理学会論文誌, 情報処理学会, 50, 2, 721-736, Feb. 2009
Scientific journal, Japanese - Reporter node determination of replicated node detection in wireless sensor networks
Yuichi Sei; Shinichi Honiden
Proceedings of the 3rd International Conference on Ubiquitous Information Management and Communication, ICUIMC'09, 566-573, 2009, Peer-reviwed, In large-scale sensor networks, sensor nodes are at high risk of being captured and compromised. Once a sensor node is compromised, all the secret keys, data, and code stored on it are exposed to the attacker. The attacker can insert arbitrary malicious code in the compromised node. Moreover, he can easily replicate such code in a large number of clones and deploy them on the network. This node replication attack can form the basis of a variety of attacks such as DoS attacks and Sybil attacks. In the related studies, as a means of detecting compromised nodes, each node reports its neighbor's ID and location to a witness node with some probability p. The value of p is determined beforehand or from only the number of neighbor nodes. In this paper, we change p according to the locations of nodes. We aim to increase the detection rate of replicated node attacks and decrease the amount of messages. Our analysis and simulations demonstrate that using our protocol in combination with the methods of other studies is more effective than using the methods of the related studies by themselves. Copyright 2009 ACM.
International conference proceedings, English - Distributed Arrays: A P2P Data Structure for Efficient Logical Arrays
Daisuke Fukuchi; Christian Sommer; Yuichi Sei; Shinichi Honiden
IEEE INFOCOM 2009 - IEEE CONFERENCE ON COMPUTER COMMUNICATIONS, VOLS 1-5, IEEE, 1458-1466, 2009, Peer-reviwed, Distributed hash tables (DHT) are used for data management in P2P environments. However, since most hash functions ignore relations between items, DHTs are not efficient for operations on related items. In this paper, we modify a DHT into a distributed array (DA) that enables efficient operations on logical arrays. The array elements of a DA are placed in a P2P overlay network according to a simple rule such that the load is balanced and the number of messages required to access elements sequentially is reduced. The number of messages required for array operations is much smaller than that for operations on DHTs. We demonstrate this theoretically and experimentally.
International conference proceedings, English - Distributed Detection of Node Replication Attacks Resilient to Many Compromised Nodes in Wireless Sensor Networks
Yuichi Sei; Shinichi Honiden
Proceedings of the 4th Annual International Conference on Wireless Internet, 28, Nov. 2008, Peer-reviwed
International conference proceedings, English - Security software engineering in Wireless Sensor Networks
Eric Platon; Yuichi Sei
Progress in Informatics, National Institute of Informatics, 5, 5, 49-64, Mar. 2008, Peer-reviwed, The engineering of security is an essential discipline in software engineering. It requires one to embrace a holistic approach, as any weakness along the engineering process of the system may lead to future security breaches. It is in general difficult to achieve and becomes particularly acute in wireless sensor networks, owing to the stringent limitations in communication and computational power. We survey the current state of the art on this topic and set forth issues that require further study. The analysis of current work covers general security issues of wireless sensor network research and discusses the present achievements for engineering security with regards to earlier surveys in this domain. We also cover security capabilities of major implementation platforms, namely TinyOS and Sun SPOTTM, and present available and required mechanisms that will become essential for software engineers.
Scientific journal, English - 無線センサーネットワークにおけるFalse Eventの検知
清雄一; 本位田真一
Lead, 情報処理学会論文誌, Information Processing Society of Japan (IPSJ), 49, 2, 628-638, Feb. 2008, Peer-reviwed, In a large scale sensor network, sensor nodes have a high risk of being captured and compromised. A compromised node can be used to generate false events. Such false events can deceive the user into wrong decisions. They can also waste a significant amount of network resources. Related works have problems; some works lose correct events stochastically and/or need to use their original routing protocols only for their methods. We propose a new method for detecting false events, which does not lose any correct events and does not specify any routing algorithms. Moreover, many security designs can address only a small threshold number of compromised nodes; the security protection completely breaks down when the threshold is exceeded. Our proposed method can achieve resiliency against an increasing number of compromised nodes. When we set the probability of losing correct events to 1%, our proposed method can detect more false events than related works do. We show this by mathematical analysis and simulations.
Scientific journal, Japanese - Energy-Efficient Event Detection in 3D Wireless Sensor Networks
Susumu Toriumi; Yuichi Sei; Shinichi Honiden
2008 1ST IFIP WIRELESS DAYS (WD), IEEE, 370-+, 2008, Peer-reviwed, Event detection techniques are crucial for environmental monitoring and object tracking applications in wireless sensor networks. Event detection requires sensor readings to he collected from multiple sensors, and as sensors have limited resources, their readings should he retrieved efficiently. Existing aggregation-based event detection methods, however, require all sensors in the network to transmit their readings, which results in high energy consumption for the network. We propose an energy-efficient event detection technique for estimating the state of the whole environment based on only some of the sensor readings and use a contour map to represent the outline of the environment. To detect events in 3D environments, we modify an existing 2D contour mapping algorithm, extending it for 3D environments. By simulation, we show how our event detection technique is more energy-efficient than existing solutions that take readings from all sensors. We evaluated our method from the point of energy efficiency and found that it improves the energy efficiency of event detection in 3D wireless sensor networks.
International conference proceedings, English - Managing difference-based objects with sub-networks in peer-to-peer environments
Fukuchi, D.; Sei, Y.; Honiden, S.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 4806 LNCS, PART 2, 1001-1010, Nov. 2007
Scientific journal, English - Resilient security for false event detection without loss of legitimate events in wireless sensor networks
Yuichi Sei; Shinichi Honiden
Lead, ON THE MOVE TO MEANINGFUL INTERNET SYSTEMS 2007: COOPLS, DOA, ODBASE, GADA, AND IS, PT 1, PROCEEDINGS, SPRINGER-VERLAG BERLIN, 4803, 454-470, Nov. 2007, Peer-reviwed, When large-scale wireless sensor networks are deployed in hostile environments, the adversary may compromise some sensor nodes and use them to generate false sensing reports or to modify the reports sent by other nodes. Such false events can cause the user to make bad decisions. They can also waste a significant amount of network resources. Unfortunately, most current security designs have drawbacks; they either require their own routing protocols to be used, or lose legitimate events stochastically and completely break down when more than a fixed threshold number of nodes are compromised. We propose a new method for detecting false events that does not suffer from these problems. When we set the probability of losing legitimate events to 1%, our proposal method can detect more false events than related method can. We demonstrate this by mathematical analysis and simulation.
International conference proceedings, English - Variable-size DBFによる分散ハッシュテーブルのトラフィック量削減
清雄一; 松崎和賢; 本位田真一
Lead, 電子情報通信学会論文誌, The Institute of Electronics, Information and Communication Engineers, 90-D, 9, 2378-2387, Sep. 2007, Peer-reviwed, ユビキタスネットワークやモバイルエージェントの基盤技術として,Peer-to-Peerシステムの一形態である分散ハッシュテーブルの研究が盛んに行われている.分散ハッシュテーブルでは,あるkeyに対応したコンテンツを早く確実に発見することが可能である.だが,複数のkeyによるAND検索においては,大量のトラヒック量が発生する.この課題に対しBloom filterというデータ構造が広く使われている.Bloom filterはFalse positive rate (FPR)というパラメータを設定する必要があるが,filterの制約により,FPRはすべてのノード間で共有する必要があった.一方,最適なFPRは各ノードによって異なる.したがって,これまでは各ノードにおける最適なFPRを実現できなかった.そこで本論文では,各ノードにおける最適なFPRを求め,各ノードがそれぞれ準最適なFPRを設定することができるような新しいfilter (variable-size DBF)を提案する.実験を行い,分散ハッシュテーブルにおけるトラヒック量を更に削減できることを示す.
Scientific journal, Japanese - Ringed filters for peer-to-peer keyword searching
Yuichi Sei; Shinichi Honiden
Lead, 16th International Conference on Computer Communications and Networks (ICCCN), IEEE, 772-779, Aug. 2007, Peer-reviwed, Distributed hash tables (DHTs) are a class of decentralized distributed systems that can efficiently search for objects desired by the user. However, a lot of communication traffic comes from multi-word searches. A lot of work has been done to reduce this traffic by using bloom filters, which are space-efficient probabilistic data structures. There are two kinds of bloom filters: fixed-size and variable-size bloom filters. We cannot use variable-size bloom filters because doing so would mean wasting time to calculating hash values. On the other hand, when using fixed-size bloom filters, all the nodes in a DHT are unable to adjust their false positive rate parameters. Therefore, the reduction of traffic is limited because the best false positive rate differs from one node to another. Moreover, in related works, the authors took only two-word searches into consideration. In this paper, we present a method for determining the best false positive rate for three- or more word searches. We also used a new filter called a ringed filter, in which each node can set the approximately best false positive rate. Experiments showed that the ringed filter was able to greatly reduce the traffic.
International conference proceedings, English - Ringed Bloom Filterによる分散ハッシュテーブルのトラフィック量削減
清雄一; 松崎和賢; 本位田真一
Lead, 情報処理学会論文誌, 48, 7, 2349-2364, Jul. 2007, Peer-reviwed, Invited
Scientific journal, Japanese - Flexible Bloom Filters for Searching Textual Objects
Yuichi Sei; Kazutaka Matsuzaki; Shinichi Honiden
International Workshop on Agents and Peer-to-Peer Computing (AP2PC), SPRINGER-VERLAG BERLIN, 5319, 110-121, May 2007, Peer-reviwed, Efficient object searching mechanisms are essential in large-scale networks. Many studies have been done on distributed hash tables (DHTs), which are a kind of peer-to-peer system. In DHT networks, we can certainly get the desired objects if they exist. However, multi-word searches generate much communication traffic. Many studies have tried to reduce this traffic by using bloom filters, which are space-efficient probabilistic data structures. In using such filters, all nodes in a DHT must share their false positive rate parameter. However, the best false positive rate differs from one node to another. In this paper, we provide a method of determining the best false positive rate, and we use a new filter called a flexible bloom filter, to which each node can set the approximately best false positive rate. Experiments showed that the flexible bloom filter was able to greatly reduce the traffic.
International conference proceedings, English - Reduction of the communication traffic for multi-word searches in DHTs
Yuichi Sei; Kazutaka Matsuzaki; Shinichi Honiden
Lead, International Conference on Intelligent Agents, Web Technologies and Internet Commerce (IAWTIC), 14-19, Nov. 2006, Peer-reviwed
International conference proceedings, English - An algorithm to reduce the communication traffic for multi-word searches in a distributed hash table
Yuichi Sei; Kazutaka Matsuzaki; Shinichi Honiden
Lead, FOURTH IFIP INTERNATIONAL CONFERENCE ON THEORETICAL COMPUTER SCIENCE - TCS 2006, SPRINGER, 209, 115-129, Aug. 2006, Peer-reviwed, In distributed hash tables, much communication traffic comes from multi-word searches. The aim of this work is to reduce the amount of traffic by using a bloom filter, which is a space-efficient probabilistic data structure used to test whether or not an element is a member of a set. However, bloom filters have a limited role if several sets have different numbers of elements. In the proposed method, extra data storage is generated when contents' keys are registered in a distributed hash table system. Accordingly, we propose a "divided bloom filter" to Solve the problem of a normal bloom filter. Using the divided bloom filter, we aim to reduce both the amount of communication traffic and the amount of data storage.
International conference proceedings, English - 分散ハッシュテーブルにおけるAND検索時のトラフィック量削減
清雄一; 松崎和賢; 本位田真一
Lead, 情報処理学会論文誌, 47, 5, 1354-1362, May 2006, Peer-reviwed
Scientific journal, Japanese
MISC
- ポジティブな単語を含んだ煽り・誹謗中傷目的のコメント検出方法の提案
佐藤豪; 清雄一; 田原康之; 大須賀 昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T1-B-3-02 - 意外性のある意見がもたらすエコーチェンバー現象の分析
中川啓; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T3-A-2-02 - 人体キーポイントを用いて映像に整合する楽器音を生成するモデルの提案
岡野日翔; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T4-A-2-04 - インターネットスラングを考慮した大規模言語モデルを用いた感情分析手法の提案
関優花; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T1-B-8-01 - スケッチによるマスク不要の表情差分生成手法の提案
佐々木嵩仁; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T4-A-7-01 - バドミントンの試合データを用いたショットの成功確率予測
美濃岡知樹; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T5-A-9-03 - ボール奪取と攻撃の防御に基づくアクションによるサッカー選手の攻守評価
前島涼弥; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T5-A-9-04 - マルチプレイヤRPGにおいて人間らしいプレイスタイルをとるゲームAIの提案
藤山仁聖; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T5-C-7-02 - ライブ配信サイトTwitchにおける人気配信者の要因の分析
佐藤大樹; 清雄一; 田原康之; 大須賀昭彦
Mar. 2024, 第16回データ工学と情報マネジメントに関するフォーラム (DEIM Forum）, T5-C-8-04, Summary national conference - 音声と3DMMに基づくマスクを除去した顔画像の推定
赤塚哲丸; 折原良平; 清雄一; 田原康之; 大須賀昭彦
Sep. 2023, 合同エージェントワークショップ＆シンポジウム (JAWS), 信学技報, 123, 190, 187-193 - 測定機器の誤差を利用した効果的な位置情報プライバシ保護手法の提案
石禾里帆; 清雄一; 田原康之; 大須賀昭彦
Sep. 2023, 合同エージェントワークショップ＆シンポジウム (JAWS), 信学技報, 123, 190, 77-82 - 擬人化タスクにおけるカラーパレットを用いた条件付き画像生成手法の挙動分析
徐江林; 折原良平; 清雄一; 田原康之; 大須賀昭彦
Sep. 2023, 合同エージェントワークショップ＆シンポジウム (JAWS), 信学技報, 123, 190, 101-108 - 化粧品レビュー文の特徴自動スコアリングを用いたレビュー文付きアイテム推薦の研究
馬場菜摘; 清雄一; 田原康之; 大須賀昭彦
Sep. 2023, 合同エージェントワークショップ＆シンポジウム (JAWS), 信学技報, 123, 190, 167-171 - 複数データソースを統合したアニメーション作品のナレッジグラフに基づく作品推薦
齋藤悠貴; 江上周作; 清雄一; 田原康之; 大須賀昭彦
Sep. 2023, 合同エージェントワークショップ＆シンポジウム (JAWS), 信学技報, 123, 190, 172-179 - 大規模条件付き画像生成モデルを用いたテキストによる出力制御可能なアイコン線画彩色
宮内洸希; 折原良平; 清雄一; 田原康之; 大須賀昭彦
Sep. 2023, 合同エージェントワークショップ＆シンポジウム (JAWS), 信学技報, 123, 190, 201-206 - Circuitry optimization and parametric study on fin-tube evaporators for maximizing COP
Cheol-Hwan Kim; Niccolo Giannetti; John Carlo; S. Garcia; Yuichi Sei; Koji Enoki; Kiyoshi Saito
Sep. 2023, JSRAE Annual Conference, English, with international co-author(s) - Evolutionary optimization of heat exchanger refrigerant circuitry
Niccolo Giannetti; Kim Cheol-Hwan; Garcia John; Carlo S; Yuichi Sei; Koji Enoki; Kiyoshi Saito
Sep. 2023, 日本冷凍空調学会年次大会, English, with international co-author(s) - 述語の意味によるクラスタリングを用いたシーングラフ生成
太田雅輝; 鵜飼孝典; 江上周作; 清雄一; 田原康之; 大須賀昭彦; 福田賢一郎
Aug. 2023, 第60回人工知能学会セマンティックウェブとオントロジー(SWO)研究会, SIG-SWO-060-02, 1-6 - ルービックキューブ求解への深層学習適用検討
水澤悟; 清雄一
Mar. 2023, 情報処理学会 全国大会 - シーングラフ生成におけるロングテール問題解決に向けたデータサンプリング手法の検討
太田 雅輝; 鵜飼 孝典; 江上 周作; 清 雄一; 田原 康之; 大須賀 昭彦; 福田 賢一郎
Mar. 2023, 第59回セマンティックウェブとオントロジー研究会, 2023, SWO-059, 09, Japanese, Summary national conference - スマートホームのセンサデータを用いた頑健なレム睡眠予測モデルの検討
津田敦哉; 清雄一; 松崎和賢
Feb. 2023, SMASH23 Winter Symposium - MLB試合データを用いた失点予測と継投計画の最適化
境田晃大; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - サッカーにおけるフィールドの位置推定モデルの提案
熊倉多香音; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - 深層強化学習を用いた文章の言い換えによる駄洒落生成モデルの提案
南智仁; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - Twitterにおけるアイコン画像と攻撃ツイートの関連性
田中智大; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - アジャイル開発プロジェクトにおける新人育成をサポートするタスク推薦システムの提案
黒木春伸; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - 深層学習を用いたスポーツ画像学習モデルによるスポーツピクトグラムの認識手法の提案
佐野景飛; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - アニメキャラクターの顔画像から全身画像への画像翻訳手法の検討
斎藤健三郎; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - 深層強化学習によるNetHack攻略に関する研究
大貫泰弘; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - 深層強化学習による人間の補助を行う格闘ゲームAIの作成
山本拓実; 清雄一; 田原康之; 大須賀昭彦
Feb. 2023, データ工学と情報マネジメントに関するフォーラム (DEIM Forum） - l-多様性を満たすためのグルーピングとダミー追加を組み合わせたアルゴリズム
大石慶一朗; 清雄一; 田原康之; 大須賀昭彦
Dec. 2022, 電子情報通信学会 人工知能と知識処理研究会, 122, 322, 80-86 - GANによるデータ拡張を用いた多様なステージ生成
高田 宗一郎; 清 雄一; 田原 康之; 大須賀 昭彦
Nov. 2022, ゲームプログラミングワークショップ, 81-87 - 2050年の知能システム
川村秀憲; 大知正直; 清雄一; 福田直樹; 横山想一郎
May 2021, 情報処理, 61, 5, 482-483, Japanese, Invited, Introduction scientific journal - Backdoor Detection Based On Network Functions For IoT Devices
依田みなみ; 櫻庭秀次; 山本純一; 清雄一; 清雄一; 田原康之; 大須賀昭彦
2020, 電子情報通信学会技術研究報告, 119, 437(ICSS2019 68-107)(Web), 0913-5685, 202002224715945078 - 低GWP冷媒を採用した次世代冷凍空調技術の実用化評価に関する研究開発(第2報)2019年度の取り組みと成果—Development of Assessment Techniques for Next-Generation Refrigerant with Low GWP Values(2nd Report)Efforts and Outcomes of FY2019
宮岡 洋一; 西山 教之; ジャンネッティ ニコロ; 鄭 宗秀; 山口 誠一; 齋藤 潔; 榎木 光治; 井上 洋平; 清 雄一; 湊 明彦; 関口 通江
日本冷凍空調学会, 2020, 日本冷凍空調学会年次大会講演論文集 Proceedings of the JSRAE Annual Conference, Japanese, 2188-5397, 40022806181 - 低GWP冷媒を採用した次世代冷凍空調技術の実用化評価に関する研究開発(第1報)次世代低GWP冷媒のサイクル性能評価研究の概要と2018年度の取り組み—Development of Assessment Techniques for Next-Generation Refrigerant with Low GWP Values(1st Report)Outline of the Research of Cycle Performance Evaluation for Next-Generation Refrigerant with Low GWP and the Approach in 2018
宮岡 洋一; ジャンネッティ ニコロ; 山口 誠一; 齋藤 潔; 榎木 光治; 井上 洋平; 清 雄一; 湊 明彦; 関口 通江
日本冷凍空調学会, 2019, 日本冷凍空調学会年次大会講演論文集 Proceedings of the JSRAE Annual Conference, Japanese, 2188-5397, 40022801443 - Construction of Anonymized Neural Network Model with Differential Privacy
清 雄一; 大須賀 昭彦
[日本ソフトウェア科学会], 09 Sep. 2015, 日本ソフトウェア科学会大会論文集, 32, 5p, Japanese, 0913-5391, 40020657336, AN10158574 - 投票行動時系列データ解析に基づく意思決定戦略に関する考察 (人工知能と知識処理)
樋口 尚吾; 折原 良平; 清 雄一
電子情報通信学会, 18 Jun. 2015, 電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 115, 97, 7-11, Japanese, 0913-5685, 40020523398, AN10013061 - Study about Mashup of the Services which are Provided by IoT Systems Built Individually
平山 秀昭; 鄭 顕志; 清 雄一; 大須賀 昭彦
Recently, M2M (Machine to Machine) or IoT (Internet of Things) is attracting attention. It connects everything in the real world to the Internet without operator's assistance. M2M or IoT has become familiar, because of miniaturization and connectivity of sensor devices. Then it is generating Big Data and becoming the driving force of realization of Smart City or Smart Community. If we can mashup the services of several M2M or IoT systems, we will be able to develop the systems which are producing greater value. But there are various problems to mashup the M2M or IoT systems. In this paper, we select an application domain of this mashup system and study its problems and solutions., Information Processing Society of Japan (IPSJ), 26 Feb. 2015, IPSJ SIG Notes, 2015, 37, 1-6, Japanese, 110009884229, AN10116224 - D-9-25 Evacuation Plan Recommendation Using Miniblogs (2) : Application of Self-Adaptive System Approach
Tahara Yasuyuki; Ohsuga Akihiko; Kawamura Takahiro; Sei Yuichi; Nakagawa Hiroyuki; Yoshioka Nobukazu; Matsumoto Kazunori; Isshiki Masao
The Institute of Electronics, Information and Communication Engineers, 24 Feb. 2015, Proceedings of the IEICE General Conference, 2015, 1, 142-142, Japanese, 110009944927, AN10471452 - Tweet classification based on expected response by user
植田 智明; 折原 良平; 清 雄一
人工知能学会, 2015, 人工知能学会全国大会論文集, 29, 1-4, Japanese, 1347-9881, 40020497720, AA11578981 - Support system for smartphone application development based on analysis of user reviews
Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
A number of smartphone applications have been developed these days. However, it is difficult to develop smartphone applications without bugs because they are used in various platforms and environments. Moreover, requirements elicitation is also difficult because various persons may use the smartphone applications. In this paper, we propose an algorithm for eliciting requirements and detecting bugs early by analyzing user reviews posted in review site of smartphone applications., Information Processing Society of Japan (IPSJ), 06 Nov. 2014, IPSJ SIG Notes, 2014, 4, 1-8, Japanese, 110009840462, AN10112981 - Data Collection for Privacy-Preserving Data Mining
清 雄一; 大須賀 昭彦
15 Oct. 2014, コンピュータセキュリティシンポジウム2014論文集, 2014, 2, 909-916, Japanese, 170000087360 - An Algorithm of k-anonymization for Location Information with Errors
SEI Yuichi; OHSUGA Akihiko
Data mining can support effective marketing or advertisement based on user' attributes such as sex, age, and current location. However, attackers can identify specific user's attributes if they know the user's location. A lot of approaches for anonymization such as k-anonymity have been proposed to tackle this problem. Existing studies, however, do not take errors of the location information into consideration. Therefore, there is a risk that a specific user's attribute can be identified by an attacker even if their information are anonymized. Moreover, the utility measure proposed in existing studies does not consider errors of the location information. We propose a novel privacy measure and a utility measure that can treat the errors of the location information and propose a method to anonymize the information of users' locations. By simulations, we prove our proposed method can improve the utility of the anonymized information and reduce the risk of the user's attribute being identified., The Institute of Electronics, Information and Communication Engineers, 28 Nov. 2013, IEICE technical report. Artificial intelligence and knowledge-based processing, 113, 332, 41-46, Japanese, 0913-5685, 110009902867, AN10013061 - L-014 Anonymized Data Collection for Attributes Containing Errors
Sei Yuichi; Ohsuga Akihiko
Forum on Information Technology, 20 Aug. 2013, 情報科学技術フォーラム講演論文集, 12, 4, 227-234, Japanese, 110009814049, AA1242354X - 不正確さを考慮した位置匿名化手法の提案
清雄一; 大須賀昭彦
03 Jul. 2013, マルチメディア、分散協調とモバイルシンポジウム2013論文集, 2013, 2044-2052, Japanese, 170000080068 - Information Sharing System on Open Education Materials for Computing Curriculum (2011 Asia Regional OpenCourseWare and Open Education Conference)
Iio Jun; Sei Yuichi; Shimizu Tomoharu
明治大学情報基盤本部, 2011, Informatics, 5, 1, 13-16, English, 1882-2908, 40019162123
Books and other publications
- Human-Centered Services Computing for Smart Cities
Yuichi Sei
Scholarly book, Contributor, Chapter 5: Privacy-Preserving Data Collection and Analysis for Smart Cities, 157-209, Springer, May 2024, Peer-reviwed - Artificial Intelligence and Blockchain Technology in Modern Telehealth Systems
Riho Isawa; Yuicih Sei; Yasuyuki Tahara; Akihiko Ohsuga; Agbotiname Lucky Imoize
Scholarly book, Contributor, Chapter 22: Location Information Privacy Protection Method Considering Human Presence Probability, The Institution of Engineering and Technology (IET), Feb. 2024, Peer-reviwed, with international co-author(s) - Handbook of Security and Privacy of AI Enabled Healthcare Systems and Internet of Medical Things
Yuichi Sei; Akihiko Ohsuga; J. Andrew Onesimu; Agbotiname Lucky Imoize
Scholarly book, English, Contributor, Chapter 10: Local Differential Privacy for Artificial Intelligence of Medical Things, CRC Press, Oct. 2023, Peer-reviwed, with international co-author(s) - Security and Privacy Schemes for Dense 6G Wireless Communication
Yuichi Sei; Akihiko Ohsuga; Agbotiname Lucky Imoize
Scholarly book, English, Contributor, Chapter 11: A Lightweight Algorithm for Detection of Fake Incident Reports in Wireless Communication Systems, 235-260, The Institution of Engineering and Technology (IET), Jul. 2023, Peer-reviwed, with international co-author(s) - Explainable Artificial Intelligence in Medical Decision Support Systems
Yuichi Sei; Akihiko Ohsuga; Agbotiname Lucky Imoize
Scholarly book, English, Contributor, Statistical Test with Differential Privacy for Medical Decision Support Systems, The Institution of Engineering and Technology (IET), Jan. 2023, Peer-reviwed, with international co-author(s) - プロジェクトをうまく進めるための17の鍵～ImprovAbilityによるプロジェクトリーダのためのプロジェクト健全化技法～
Yuichi Sei; Yasuhiro Kikushima; Yasushi Ishigai; Jan Pries-Heje; Jørn Johansen
General book, Japanese, Joint work, Ohmsha, Aug. 2017, with international co-author(s), 9784274700002 - CoBRA法入門 「勘」を見える化する見積り手法
CoBRA研究会
Scholarly book, Japanese, Contributor, オーム社, Apr. 2011, 9784274503351 - Designing Solutions-Based Ubiquitous and Pervasive Computing: New Issues and Trends
Kenji Tei; Shunichiro Suenaga; Yoshiyuki Nakamura; Yuichi Sei; Hikotoshi Nakazato; Yoichi Kaneki; Nobukazu Yoshioka; Yoshiaki Fukazawa; Shinichi Honiden
Scholarly book, English, Contributor, Chapter 11: XAC Project: Towards a Middleware for Open Wireless Sensor Networks, IGI Global, Apr. 2010, Peer-reviwed
Lectures, oral presentations, etc.
- AI技術の熱・空調への応用
清雄一
Public discourse, 次世代ヒートポンプ研究戦略研究コンソーシアム, Invited
Apr. 2024 - Privacy-Preserving Collaborative Data Collection and Analysis with Many Missing Values
Yuichi Sei
Public discourse, 電気通信普及財団賞贈呈式, Invited, with international co-author(s)
21 Mar. 2024 - Analysis of the Echo Chamber Caused by Unexpected Opinions
Akira Nakagawa; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Oral presentation, 16th International Conference on Agents and Artificial Intelligence (ICAART), Peer-reviewed, International conference
Feb. 2024 - Enhance Data Usefulness in Privacy Protection Under Considering IoT Measurement Error
Riho Isawa; Yuicih Sei; Yasuyuki Tahara; Akihiko Ohsuga
Oral presentation, 16th International Conference on Agents and Artificial Intelligence (ICAART), Peer-reviewed, International conference
Feb. 2024 - Reconstruction of Facial Geometry from Face-Masked Images Using Voice Cues
Tetsumaru Akatsuka; Ryohei Orihara; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
Oral presentation, 16th International Conference on Agents and Artificial Intelligence (ICAART), Peer-reviewed, International conference
Feb. 2024 - AIを用いた伝熱性能予測
清雄一
早稲田大学 持続的環境エネルギー社会共創研究機構 研究所間交流会
15 Sep. 2023 - Investigation of Pitch Risk Model Using Goal and Assist Information and Pitch Control
Yuya Jingushi; Yuichi Sei; Yasuyuki Tahara; Akihiko Ohsuga
International Conference on Agents and Artificial Intelligence (ICAART), Peer-reviewed
Feb. 2023 - Machine Learning on Differentially Private Data
Yuichi Sei
Keynote oral presentation, International Conference on AI and Machine Learning, Invited
Oct. 2022
26 Oct. 2022- 27 Oct. 2022 - Web/IoT横断的プライバシ保護データ解析基盤
清雄一
Invited oral presentation, Japanese, SMASH20 Winter Symposium, Invited, Domestic conference
Dec. 2020 - AI技術の活用事例～管内沸騰熱伝達率の整理～
清雄一; 榎木光治
Invited oral presentation, Japanese, 技術セミナー（西日本地区）AI技術の基礎と冷凍空調分野への応用に向けて, Invited, 日本冷凍空調学会, Domestic conference
Feb. 2020 - Privacy-Preserving IoT Data Mining
Yuichi Sei
Keynote oral presentation, English, Conference on Intelligent Computing, Communication & Applied Technologies (CICCAT), Invited, International conference
Dec. 2019 - Privacy-preserving IoT Data Mining
Yuichi Sei
Keynote oral presentation, English, Conference on Intelligent Computing, Communication & Applied Technologies, Invited
Dec. 2019 - AI技術の概観と熱交換器への応用
清雄一
Invited oral presentation, Japanese, 日本冷凍空調学会「環境変化に対応するための先進熱交換技術に関する調査研究」委員会, Invited, Domestic conference
Mar. 2019 - Privacy-Preserving Data Collection and Sharing for Big Data
Yuichi Sei; Akihiko Ohsuga
Invited oral presentation, English, International Conference for Top and Emerging Computer Scientists (IC-TECS), Invited, International conference
Dec. 2017 - Anonymized Data Collection Based on Randomized Multiple Values
Yuichi Sei; Akihiko Ohsuga
Invited oral presentation, English, The Seventh Symposium on Biometrics, Recognition and Authentication, Invited, Domestic conference
Nov. 2017 - Design based on scenario to consider the improvement of reliability in a conceptual design stage
AOYAMA Kazuhiro; EZOE Tomosuke; KOGA Tsuyoshi; SEI Yuuichi
Oral presentation, Design & Systems Conference
Mar. 2004
Courses
- Physics Laboratory
2023 - Present
The University of Electro-Communications, Undergraduate liberal arts - Physics Laboratory
2023 - Present
The University of Electro-Communications, Undergraduate liberal arts - Advanced System Design III
2021 - Present
The University of Electro-Communications, Postgraduate courses - Ubiquitous Networks
2018 - Present
The University of Electro-Communications, Undergraduate special subjects - Media Computing System
2017 - Present
The University of Electro-Communications, Undergraduate special subjects - Exercise in Informatics I
2016 - Present
The University of Electro-Communications, Undergraduate liberal arts - Physics Laboratory
2017 - 2023
The University of Electro-Communications - Advanced Mathematical Sciences
2019 - 2020
Meiji University, Postgraduate courses - Machine Learning
2018 - 2020
Smart SE, Waseda University - Advanced System Design I
2018 - 2020
The University of Electro-Communications, Postgraduate courses - Fundamentals of Social Intelligence Informatics II
The University of Electro-Communications
Affiliated academic society
Works
Research Themes
- 生成AI遍在社会におけるプライバシ保護基盤の創成
清 雄一; 石川冬樹; 松崎 和賢; 田原康之; 大須賀昭彦
日本学術振興会, 科学研究費助成事業, 電気通信大学, 基盤研究(A), Principal investigator, 24H00714
Apr. 2024 - Mar. 2028 - SNS・IoT・オープンデータ融合マイニングによる施策に対する人々の行動変化予測
大須賀昭彦; 田原康之; 清雄一; 吉岡信和; 鄭顕志; 江上周作
日本学術振興会, 科学研究費助成事業, 基盤研究(B), Coinvestigator, 23H03688
Apr. 2023 - Mar. 2027 - シャッフル差分プライバシーの安全性と有用性の向上に関する研究
統計数理研究所, 公募型共同利用, Principal investigator
Apr. 2024 - Mar. 2025 - Privacy Protection Infrastructure in a Ubiquitous Machine Learning Society
清 雄一; 大須賀 昭彦; 田原 康之; 松崎 和賢
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific, The University of Electro-Communications, Grant-in-Aid for Scientific Research (B), Principal investigator, 21H03496
Apr. 2021 - Mar. 2025 - 低GWP冷媒の熱伝達特性の予測技術に関する研究
宇宙航空研究開発機構（JAXA）, Coinvestigator
Jul. 2023 - Mar. 2024 - 社会課題解決に向けたパーソナルデータ活用機械学習モデルとプライバシ
公益財団法人住友電工グループ社会貢献基金, 研究助成, Principal investigator
Oct. 2022 - Mar. 2024 - Research on Privacy-Preserving Data Analysis Platform for Web and IoT
Japan Science and Technology Agency (JST), Precursory Research for Embryonic Science and Technology (PRESTO), Principal investigator
Oct. 2019 - Mar. 2023 - 省エネ化・低温室効果を達成できる次世代冷凍空調技術の最適化及び評価手法の開発
NEDO（国立研究開発法人 新エネルギー・産業技術総合開発機構）
Aug. 2018 - Mar. 2023 - A Study on Locally Private Algorithms for Large-Scale Personal Data
Murakami Takao
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, National Institute of Advanced Industrial Science and Technology, Grant-in-Aid for Scientific Research (B), In this work, we studied locally private algorithms for large-scale personal data, such as time-series data (e.g., location traces) and social graph data. Specifically, we proposed a locally private algorithm based on LSH (Locality Sensitive Hashing), a location trace synthesizer, and graph LDP (Local Differential Privacy) algorithms with utility guarantees. We also proposed a privacy notion called ULDP (Utility-Optimized LDP), which provides privacy guarantees equivalent to LDP for only sensitive data., 19H04113
Apr. 2019 - Mar. 2022 - Social and physical sensor fusion mining infrastructure to understand behavioral intentions
Ohsuga Akihiko
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, The University of Electro-Communications, Grant-in-Aid for Scientific Research (B), The results of this research are as follows. (1) Development of an algorithm to remove universal patterns and noise from anonymized data, (2) Development of a hierarchical deep learning algorithm that considers the granularity and accuracy of input values for individual social data, (3) Construction of heterogeneous Linked Data to link anonymized physical data and social data that are likely to be information of the same person. (4) Extraction of actions that increase or decrease the likelihood of goal attainment based on goal-directed actions, and (5) Development of an example application that recommends optimal actions based on social media data and physical sensor data., 18H03340
Apr. 2018 - Mar. 2022 - Research on autonomous cooperative self-adaptation mechanisms and formal verification of them
Tahara Yasuyuki
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, The University of Electro-Communications, Grant-in-Aid for Scientific Research (B), The results of this research are as follows: (1) development of a cooperative method for other devices to notice environmental changes that cannot be detected by individual devices after careful consultation among them, (2) development of a method for cooperative adaptation to environmental changes that cannot be handled by individual devices, (3) development of a method for verifying whether cooperative behavior works correctly by applying formal verification, (4) development of a mechanism for preventing privacy violations in IoT systems, including those caused by self-adaptive mechanisms for cooperative behavior,and (5) Development, experimentation, and evaluation of autonomous cooperative self-adaptive middleware and applications for IoT that integrate the mechanisms and methods in (1) through (4)., 18H03229
Apr. 2018 - Mar. 2022 - Flexible framework of privacy-preserving IoT data analysis
Sei Yuichi
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, The University of Electro-Communications, Grant-in-Aid for Challenging Research (Exploratory), Poople and organizations have been building and promoting new services that utilize a cross-section of Internet of Things (IoT) data. While this will greatly improve people's convenience, it will also make it unpredictable from where personal privacy information will leak out, and the construction of a common and robust framework to protect privacy will be an important issue. In this study, we built a privacy protection framework for IoT data with errors and inadequacies., 18K19835
Jun. 2018 - Mar. 2021 - Identification of privacy risk in an Open Data society
Sei Yuichi
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, The University of Electro-Communications, Grant-in-Aid for Young Scientists (A), Social media users are often anonymous and may also use pseudonyms to hide sensitive information such as disease. However, when linked to open data, there is a possibility that individuals may be identified or their personal attribute values may be leaked. In Japan, Act on the Protection of Personal Information and the Open Data Strategy have promoted the release and sharing of anonymized personal information and statistical data. In this research, we will study the risk of privacy information being inferred in conjunction with open data., 17H04705
Apr. 2017 - Mar. 2021 - 第2期enPiT (Education Network for Practical Information Technologies: 成長分野を支える情報技術人材の育成拠点の形成)
文部科学省
Sep. 2016 - Mar. 2021 - 必要な知識を自動的に獲得して結論を導出する推論フレームワークの研究
中島記念国際交流財団
Apr. 2019 - Mar. 2020 - 河川水位データの異常値検出
日本工営株式会社, Principal investigator, Domestic joint research
Apr. 2018 - Mar. 2019 - Research on big data magnifying glasses for data processing considering urgency
Akihiko Ohsuga; Yasuyuki Tahara; Yuichi Sei
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, The University of Electro-Communications, Grant-in-Aid for Challenging Exploratory Research, In this research, we developed a big data processing middleware that enables the construction of systems that adapt to changing urgency levels by changing the degree of abstraction of the data and the degree of protection of the privacy according to the local degree of urgency determined automatically, as if you moved the magnifying glass up and down and left and right to look at the data. For this purpose, we established data abstraction level adjustment method and privacy protection degree adjustment method, constructed an architecture to switch these layers according to the level of urgency, integrated it as middleware with an interface that imitates the operation of the magnifying glass to handle the change of level of urgency, developed a big data application example, evaluated it experimentally, and published the results., 16K12411
Apr. 2016 - Mar. 2019 - 年金給付システムに係る法令工学文献調査
株式会社三菱総合研究所, Principal investigator, Research commissioned by a company
Apr. 2017 - Mar. 2018 - 匿名化アルゴリズムおよび匿名化データ分析手法に関する研究２
株式会社三菱総合研究所, Principal investigator, Domestic joint research
Oct. 2016 - Mar. 2017 - 超上流を重視したプロジェクトマネジメント改善フレームワークの調査研究とその拡張
情報科学国際交流財団産学戦略的研究フォーラム, 研究助成
Jun. 2016 - Mar. 2017 - プライバシ保護ディープラーニングのためのニューラルネットワークモデル構築手法の提案
公益財団法人電気通信普及財団, 研究助成
Apr. 2016 - Mar. 2017 - Anonymization of Personal Data in Ubiquitous Computing
Sei Yuichi
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, The University of Electro-Communications, Grant-in-Aid for Young Scientists (B), Principal investigator, In recent years, numerous organizations have begun to provide services that collect large amounts of personal information. This personal information can be shared with other organizations so that they can subsequently create new services. Owing to the development of ubiquitous computing and sensing technologies, numerous research methods for crowdsensing have been proposed to collect and analyze sensed environmental information from mobile phone users. Our contributions are as follows: (1) we propose a simple but effective general anonymization algorithm for large databases, (2) we propose a novel privacy metric and a utility metric that can treat the location error and propose an efficient anonymization algorithm for the proposed metrics, and (3) we propose S2M and S2Mb, both of which can make a better trade-off between privacy and utility in crowdsensing., 26870201
Apr. 2014 - Mar. 2017 - Efficient runtime formal verification of aspect-oriented models@run.time systems
Tahara Yasuyuki
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C), The University of Electro-Communications, Grant-in-Aid for Scientific Research (C), We developed a code generation system that provides an abstraction / refinement relation between the model and the source code based on the framework of the formal model of aspect-oriented models@run.time system proposed by us. We extended the framework of the formal model so that it can deal with the behavior in which aspects are woven at runtime for adaptation, by adding, as new axioms, axiom changing rules that formalize aspect weaving using reflection. We developed a prototype of verification and adaptation integrating the code generation system and the verification system based on the design of the prototype of the verification / adaptation system. We also implemented, conducted experiments of, and evaluated the example web / ubiquitous application., 26330081
Apr. 2014 - Mar. 2017 - Research on anonymization algorithms and methods for analyzing anonymized data
株式会社三菱総合研究所, Mitsubishi Research Institute, Principal investigator, Domestic joint research
Apr. 2016 - Sep. 2016 - 無線センサネットワークの信頼性向上に向けた自己適応フレームワーク
国立情報学研究所
Oct. 2014 - Mar. 2015 - Research on Anonymization of User Information in Ubiquitous Computing
Yuichi Sei
Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research Grant-in-Aid, The University of Electro-Communications, Grant-in-Aid for Research Activity Start-up, Principal investigator, ユーザ属性項目数の増加及びサーバへの攻撃の多様化に対応するため，サーバが信頼できない状況においてもユーザ項目を匿名化可能な手法を提案し，ユーザ属性項目数が多数ある場合に特に有効な手法へと拡張を行った．具体的には，Negative Survey及びRandomized Responseで利用される確率行列を改善することにより，プライバシ保護レベルを一定に保ちつつ匿名化後のデータ誤差を50%から97%程度削減することができた．さらに，属性項目数が多い場合においては匿名化後のデータ誤差を85%から99.99%程度削減することができた． 上記に加え，誤差を含む位置情報を対象とした匿名化手法を提案した．既存手法と比べ，プライバシ保護レベルを10%程度向上させるとともに，匿名化後のデータの有効性を20%程度向上させることができた． また，ユーザごとに要求するプライバシ保護レベルが異なるとき，その差異を統計的に扱うことによって従来よりも匿名化後のデータの有効性をさらに70%程度向上させた． さらに，ユーザ属性を取得するための無線センサネットワークにおいて，ユーザ属性を不正に取得する脅威に対応するための手法を提案した．, 25880009
Aug. 2013 - Mar. 2015
Industrial Property Rights
- Privacy-preserving data delivery system
Patent right, 清雄一, 奥村拓史, 大須賀昭彦, 特願2016-239460, Date applied: 09 Dec. 2016, The University of Electro-Communications, Mitsubishi Research Institute, 特開2018-97467, Date announced: 21 Jun. 2018, 特許6835559, Date registered: 08 Feb. 2021, Date issued: 24 Feb. 2021 - Moving image distribution system, moving image dividing system, moving image distribution program, moving image dividing program, and recording medium storing moving image distribution program and/or moving image dividing program
Patent right, Nakamura, J, Saito, H, Hiramoto, K, Sei, Y, Kozhevnikov, S, Matsuda, T, Togo, J, Ono, S, Kato, S, Arakawa, Y, 特願2005-513408, Date applied: 27 Jul. 2004, WO2005-022912, Date announced: 02 Nov. 2006, 特許第3858048号, Date issued: 22 Sep. 2006
Media Coverage
- AI・ゲノムで個人特定 捜査利用も、情報保護課題
日経産業新聞, 朝刊7面, Paper
Mar. 2024 - Enhance data usefulness in privacy protection under considering IoT measurement error
EurekAlert, Internet
Feb. 2024 - マスクに隠れた顔をAIで再現
日本経済新聞, 朝刊16面, Paper
Jan. 2024 - Study addresses privacy-preserving collaborative data collection and analysis with many missing values
Myself, Tech Xplore, Internet
Jul. 2023 - Individual re-identification from incomplete datasets protected by differential privacy
EurekAlert, Internet
Jul. 2022 - Web Internet of Things for analyzing data while protecting privacy
PR Newswire, Others
Mar. 2021 - 熱交換器設計、ＡＩで効率化 電通大、沸騰熱伝達の予測技術
日刊工業新聞, Paper
Jan. 2018 - 電気通信大と早大、深層学習により高精度な沸騰熱伝達予測モデルを構築
日本経済新聞, Paper
Dec. 2017 - 秘匿性と実用性 両立 電通大 ビッグデータ解析技術
Other than myself, 日刊工業新聞, 朝刊1面トップ, Paper
Jul. 2017
Academic Contribution Activities
- International Joint Conference on Artificial Intelligence (IJCAI)
Academic society etc, Peer review, Jan. 2022 - Present - 22nd International Conference on Practical applications of Agents and Multi-Agent Systems (PAAMS)
Academic society etc, Planning etc, Apr. 2024 - Aug. 2024, True - 15th International Conference on Smart Computing and Artificial Intelligence
Academic society etc, Planning etc, Jan. 2024 - Jul. 2024, True - 18th International Conference on E-Service and Knowledge Management
Academic society etc, Planning etc, Jan. 2024 - Jul. 2024, True - 情報処理学会全国大会
Competition etc, Panel chair etc, Mar. 2024 - Mar. 2024 - 情報・システムソサイエティ学術奨励賞選定委員会
Academic society etc, Review, 電子情報通信学会, Aug. 2023 - Mar. 2024 - IEICE Monograph編集委員会，編集委員
Academic society etc, Planning etc, 電子情報通信学会, May 2023 - Mar. 2024 - Symposium on Multi Agent Systems for Harmonization (SMASH)
Academic society etc, Planning etc, Feb. 2024 - Feb. 2024 - 16th International Conference on Agents and Artificial Intelligence (ICAART)
Academic society etc, Panel chair etc, Feb. 2024 - Feb. 2024, True - ACM SIGSPATIAL International Workshop on Geo-Privacy and Data Utility for Smart Societies (GeoPrivacy)
Academic society etc, Planning etc, Jul. 2023 - Dec. 2023, Hamburg, Germany, True - コンピュータセキュリティシンポジウム2023
Planning etc, Jun. 2023 - Dec. 2023 - 14th International Conference on Smart Computing and Artificial Intelligence
Academic society etc, Planning etc, Jun. 2023 - Dec. 2023, True - 17th International Conference on E-Service and Knowledge Management
Academic society etc, Planning etc, Jun. 2023 - Dec. 2023, True - International Conference on Practical applications of Agents and Multi-Agent System (PAAMS)
Competition etc, Peer review, Apr. 2023 - Oct. 2023, True - 合同エージェントワークショップ＆シンポジウム2023 (JAWS2023)
Competition etc, Panel chair etc, 電子情報通信学会 人工知能と知識処理専門委員会（IEICE SIG-AI） 情報処理学会 知能システム研究会（IPSJ SIG-ICS） マルチエージェントと協調計算研究会（JSSST MACC） 人工知能学会 データ指向構成マイニングとシミュレーション研究会（JSAI SIG-DOCMAS） IEEE Computer Society Tokyo/Japan Joint Chapter, 12 Sep. 2023 - 14 Sep. 2023 - FIT2023
Competition etc, Planning etc, 一般社団法人電子情報通信学会 情報・システムソサイエティ（ISS） ヒューマンコミュニケーショングループ（HCG） 一般社団法人情報処理学会（FIT2023幹事学会）, 06 Sep. 2023 - 08 Sep. 2023 - 13th International Conference on Smart Computing and Artificial Intelligence
Planning etc, Apr. 2023 - Jul. 2023, True - 16th International Conference on E-Service and Knowledge Management
Planning etc, Apr. 2023 - Jul. 2023, True - 情報処理学会全国大会
Competition etc, Panel chair etc, Mar. 2023 - Mar. 2023 - 電子情報通信学会総合大会
Competition etc, Panel chair etc, Mar. 2023 - Mar. 2023 - SMSH23 Winter Symposium
Competition etc, Planning etc, Dec. 2022 - Feb. 2023 - FIT 2022
Planning etc, Jan. 2022 - Oct. 2022 - SMASH22 Summer Symposium
Academic society etc, Planning etc, Jun. 2022 - Sep. 2022 - International Conference on E-Service and Knowledge Management (ESKM)
Academic society etc, Peer review, Apr. 2022 - Jul. 2022 - International Conference on Smart Computing and Artificial Intelligence (SCAI)
Academic society etc, Others, Apr. 2022 - Jul. 2022 - International Joint Conference on Artificial Intelligence (IJCAI)
Academic society etc, Others, 2021 - IEEE International Conference on Big Data Analytics (ICBDA)
Peer review etc, Peer review, 2020 - IEEE Transactions on Knowledge and Data Engineering
Peer review etc, Peer review, 2020 - Symposium of Multi Agent Systems for Harmonization 2020 (SMASH20)
Academic society etc, Others, 2020 - 31st IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC): Track2: Networking and MAC
Academic society etc, Others, 2020