Towards a Privacy-Aware Recommendation System for Manga Reading Application

• Item type: Symposium(1)
• 公開日: 2019-10-14
• タイトル: Towards a Privacy-Aware Recommendation System for Manga Reading Application
• タイトル / 言語: en
• タイトル: Towards a Privacy-Aware Recommendation System for Manga Reading Application
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: Privacy，Recommendation System，Machine Learning
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_5794
• 資源タイプ: conference paper
• 著者所属: Graduate School of Information Science and Technology The University of Tokyo
• 著者所属: Graduate School of Information Science and Technology The University of Tokyo
• 著者所属: Graduate School of Information Science and Technology The University of Tokyo
• 著者所属(英): en
• 著者所属(英): Graduate School of Information Science and Technology The University of Tokyo
• 著者所属(英): en
• 著者所属(英): Graduate School of Information Science and Technology The University of Tokyo
• 著者所属(英): en
• 著者所属(英): Graduate School of Information Science and Technology The University of Tokyo
• 著者名: Mhd, Irvan; Toshiyuki, Nakata; Rie, Shigetomi Yamaguchi
• 著者名(英): Mhd, Irvan; Toshiyuki, Nakata; Rie, Shigetomi Yamaguchi
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Reading habits can potentially reveal many characteristics of the readers that they likely want to keep private to themselves. Furthermore, many on-demand reading applications nowadays are being deployed as smartphone applications, allowing even more delicate data to be detected and shared away from the phone itself. These information are useful to feed into centralized machine learning programs to, for example, recommend interesting contents. This paper argues that it is possible to build reliable recommendation systems without gathering those data into a centralized place, beyond the users' control. We propose a privacy-preserving machine learning approach that can be applied to recommendation systems. This approach is tested on a manga reading application dataset to demonstrate its usefulness in real world usage.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Reading habits can potentially reveal many characteristics of the readers that they likely want to keep private to themselves. Furthermore, many on-demand reading applications nowadays are being deployed as smartphone applications, allowing even more delicate data to be detected and shared away from the phone itself. These information are useful to feed into centralized machine learning programs to, for example, recommend interesting contents. This paper argues that it is possible to build reliable recommendation systems without gathering those data into a centralized place, beyond the users' control. We propose a privacy-preserving machine learning approach that can be applied to ecommendation systems. This approach is tested on a manga reading application dataset to demonstrate its usefulness in real world usage.
• 書誌レコードID / 識別子タイプ: NCID
• 書誌レコードID / 関連識別子: ISSN 1882-0840
• 書誌情報: コンピュータセキュリティシンポジウム2019論文集; 巻 2019, p. 1493-1496, 発行日 2019-10-14
• 出版者 / 言語: ja
• 出版者: 情報処理学会