Reducing hubness: A cause of vulnerability in recommender systems

Kazuo Hara, Ikumi Suzuki, Kei Kobayashi, Kenji Fukumizu

研究成果: Conference contribution

5 被引用数 (Scopus)

抄録

It is known that memory-based collaborative filtering systems are vulnerable to shilling attacks. In this paper, we demonstrate that hubness, which occurs in high dimensional data, is exploited by the attacks. Hence we explore methods for reducing hubness in user-response data to make these systems robust against attacks. Using the MovieLens dataset, we empirically show that the two methods for reducing hubness by transforming a similarity matrix|(i) centering and (ii) conversion to a commute time kernel|can thwart attacks without degrading the recommendation performance.

本文言語English
ホスト出版物のタイトルSIGIR 2015 - Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval
出版社Association for Computing Machinery, Inc
ページ815-818
ページ数4
ISBN(電子版)9781450336215
DOI
出版ステータスPublished - 2015 8月 9
外部発表はい
イベント38th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2015 - Santiago, Chile
継続期間: 2015 8月 92015 8月 13

Other

Other38th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2015
国/地域Chile
CitySantiago
Period15/8/915/8/13

ASJC Scopus subject areas

  • 情報システム
  • ソフトウェア

フィンガープリント

「Reducing hubness: A cause of vulnerability in recommender systems」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル