M-BPR: A novel approach to improving BPR for recommendation with multi-type pair-wise preferences

Yeon Chang Lee, Taeho Kim, Jaeho Choi, Xiangnan He, Sang Wook Kim

Research output: Contribution to journalArticle

Abstract

In this paper, we examine the two assumptions of the Bayesian personalized ranking (BPR), a well-known pair-wise method for one-class collaborative filtering (OCCF): (1) a user with the same degree of negative preferences for all her unrated items; and (2) a user always preferring her rated items to all her unrated items. We claim that (A1) and (A2) cause recommendation errors because they do not always hold in practice. To address these problems, we first define fine-grained multi-type pair-wise preferences (PPs), which are more sophisticated than the single-type PP used in BPR. Then, we propose a novel pair-wise approach called M-BPR, which exploits multi-type PPs together in learning users’ more detailed preferences. Furthermore, we refine M-BPR by employing the concept of item groups to reduce the uncertainty of a user's a single item-level preference. Through extensive experiments using four real-life datasets, we demonstrate that our approach addresses the problems of the original BPR effectively and also outperforms seven state-of-the-art OCCF (i.e., four pair-wise and three point-wise) methods significantly.

Original languageEnglish
Pages (from-to)255-270
Number of pages16
JournalInformation Sciences
Volume547
DOIs
StatePublished - 2021 Feb 8

Keywords

  • Bayesian personalized ranking
  • One-class collaborative filtering
  • Pair-wise preferences
  • Recommender systems

Fingerprint Dive into the research topics of 'M-BPR: A novel approach to improving BPR for recommendation with multi-type pair-wise preferences'. Together they form a unique fingerprint.

Cite this