Disentangled Negative Sampling for Collaborative Filtering

Riwei Lai, Li Chen, Yuhan Zhao, Rui Chen*, Qilong Han*

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

3 Citations (Scopus)

Abstract

Negative sampling is essential for implicit collaborative filtering to generate negative samples from massive unlabeled data. Unlike existing strategies that consider items as a whole when selecting negative items, we argue that normally user interactions are mainly driven by some relevant, but not all, factors of items, leading to a new direction of negative sampling. In this paper, we introduce a novel disentangled negative sampling (DENS) method. We first disentangle the relevant and irrelevant factors of positive and negative items using a hierarchical gating module. Next, we design a factor-aware sampling strategy to identify the best negative samples by contrasting the relevant factors while keeping irrelevant factors similar. To ensure the credibility of the disentanglement, we propose to adopt contrastive learning and introduce four pairwise contrastive tasks, which enable to learn better disentangled representations of the relevant and irrelevant factors and remove the dependency on ground truth. Extensive experiments on five real-world datasets demonstrate the superiority of DENS against several state-of-the-art competitors, achieving over 7% improvement over the strongest baseline in terms of Recall@20 and NDCG@20. Our code is publically available at https://github.com/Riwei-HEU/DENS .

Original languageEnglish
Title of host publicationWSDM '23: Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining
PublisherAssociation for Computing Machinery (ACM)
Pages96-104
Number of pages9
ISBN (Electronic)9781450394079
DOIs
Publication statusPublished - 27 Feb 2023
Event16th ACM International Conference on Web Search and Data Mining, WSDM 2023 - , Singapore
Duration: 27 Feb 20233 Mar 2023
https://dl.acm.org/doi/proceedings/10.1145/3539597

Publication series

NameWSDM - Proceedings of the ACM International Conference on Web Search and Data Mining

Conference

Conference16th ACM International Conference on Web Search and Data Mining, WSDM 2023
Country/TerritorySingapore
Period27/02/233/03/23
Internet address

Scopus Subject Areas

  • Computer Networks and Communications
  • Computer Science Applications
  • Software

User-Defined Keywords

  • contrastive learning
  • disentanglement learning
  • negative sampling

Fingerprint

Dive into the research topics of 'Disentangled Negative Sampling for Collaborative Filtering'. Together they form a unique fingerprint.

Cite this