Dual-regularized one-class collaborative filtering with implicit feedback

Yuan Yao, Hanghang Tong, Guo Yan, Feng Xu, Xiang Zhang, Boleslaw K. Szymanski, Jian Lu

Research output: Contribution to journalArticle

Abstract

Collaborative filtering plays a central role in many recommender systems. While most of the existing collaborative filtering methods are proposed for the explicit, multi-class settings (e.g., 1-5 stars in movie recommendation), many real-world applications actually belong to the one-class setting where user feedback is implicitly expressed (e.g., views in news recommendation and video recommendation). In this article, we propose dual-regularized one-class collaborative filtering models for implicit feedback. In particular, by dividing existing methods into point-wise class and pair-wise class, we first propose a point-wise model by integrating two existing methods and further exploiting the side information from both users and items. Next, we propose to add dual regularization into an existing pair-wise method with a different treatment of the side information. We also propose efficient algorithms to solve the proposed models. Extensive experimental evaluations on three real data sets demonstrate the effectiveness and efficiency of the proposed methods.

Original languageEnglish (US)
Pages (from-to)1099-1129
Number of pages31
JournalWorld Wide Web
Volume22
Issue number3
DOIs
StatePublished - May 15 2019

Fingerprint

Collaborative filtering
Feedback
Recommender systems
Stars

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Yao, Y., Tong, H., Yan, G., Xu, F., Zhang, X., Szymanski, B. K., & Lu, J. (2019). Dual-regularized one-class collaborative filtering with implicit feedback. World Wide Web, 22(3), 1099-1129. https://doi.org/10.1007/s11280-018-0574-1
Yao, Yuan ; Tong, Hanghang ; Yan, Guo ; Xu, Feng ; Zhang, Xiang ; Szymanski, Boleslaw K. ; Lu, Jian. / Dual-regularized one-class collaborative filtering with implicit feedback. In: World Wide Web. 2019 ; Vol. 22, No. 3. pp. 1099-1129.
@article{57535962000c4c0d9896e3f396f44336,
title = "Dual-regularized one-class collaborative filtering with implicit feedback",
abstract = "Collaborative filtering plays a central role in many recommender systems. While most of the existing collaborative filtering methods are proposed for the explicit, multi-class settings (e.g., 1-5 stars in movie recommendation), many real-world applications actually belong to the one-class setting where user feedback is implicitly expressed (e.g., views in news recommendation and video recommendation). In this article, we propose dual-regularized one-class collaborative filtering models for implicit feedback. In particular, by dividing existing methods into point-wise class and pair-wise class, we first propose a point-wise model by integrating two existing methods and further exploiting the side information from both users and items. Next, we propose to add dual regularization into an existing pair-wise method with a different treatment of the side information. We also propose efficient algorithms to solve the proposed models. Extensive experimental evaluations on three real data sets demonstrate the effectiveness and efficiency of the proposed methods.",
author = "Yuan Yao and Hanghang Tong and Guo Yan and Feng Xu and Xiang Zhang and Szymanski, {Boleslaw K.} and Jian Lu",
year = "2019",
month = "5",
day = "15",
doi = "10.1007/s11280-018-0574-1",
language = "English (US)",
volume = "22",
pages = "1099--1129",
journal = "World Wide Web",
issn = "1386-145X",
publisher = "Springer New York",
number = "3",

}

Yao, Y, Tong, H, Yan, G, Xu, F, Zhang, X, Szymanski, BK & Lu, J 2019, 'Dual-regularized one-class collaborative filtering with implicit feedback', World Wide Web, vol. 22, no. 3, pp. 1099-1129. https://doi.org/10.1007/s11280-018-0574-1

Dual-regularized one-class collaborative filtering with implicit feedback. / Yao, Yuan; Tong, Hanghang; Yan, Guo; Xu, Feng; Zhang, Xiang; Szymanski, Boleslaw K.; Lu, Jian.

In: World Wide Web, Vol. 22, No. 3, 15.05.2019, p. 1099-1129.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Dual-regularized one-class collaborative filtering with implicit feedback

AU - Yao, Yuan

AU - Tong, Hanghang

AU - Yan, Guo

AU - Xu, Feng

AU - Zhang, Xiang

AU - Szymanski, Boleslaw K.

AU - Lu, Jian

PY - 2019/5/15

Y1 - 2019/5/15

N2 - Collaborative filtering plays a central role in many recommender systems. While most of the existing collaborative filtering methods are proposed for the explicit, multi-class settings (e.g., 1-5 stars in movie recommendation), many real-world applications actually belong to the one-class setting where user feedback is implicitly expressed (e.g., views in news recommendation and video recommendation). In this article, we propose dual-regularized one-class collaborative filtering models for implicit feedback. In particular, by dividing existing methods into point-wise class and pair-wise class, we first propose a point-wise model by integrating two existing methods and further exploiting the side information from both users and items. Next, we propose to add dual regularization into an existing pair-wise method with a different treatment of the side information. We also propose efficient algorithms to solve the proposed models. Extensive experimental evaluations on three real data sets demonstrate the effectiveness and efficiency of the proposed methods.

AB - Collaborative filtering plays a central role in many recommender systems. While most of the existing collaborative filtering methods are proposed for the explicit, multi-class settings (e.g., 1-5 stars in movie recommendation), many real-world applications actually belong to the one-class setting where user feedback is implicitly expressed (e.g., views in news recommendation and video recommendation). In this article, we propose dual-regularized one-class collaborative filtering models for implicit feedback. In particular, by dividing existing methods into point-wise class and pair-wise class, we first propose a point-wise model by integrating two existing methods and further exploiting the side information from both users and items. Next, we propose to add dual regularization into an existing pair-wise method with a different treatment of the side information. We also propose efficient algorithms to solve the proposed models. Extensive experimental evaluations on three real data sets demonstrate the effectiveness and efficiency of the proposed methods.

UR - http://www.scopus.com/inward/record.url?scp=85046149529&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046149529&partnerID=8YFLogxK

U2 - 10.1007/s11280-018-0574-1

DO - 10.1007/s11280-018-0574-1

M3 - Article

AN - SCOPUS:85046149529

VL - 22

SP - 1099

EP - 1129

JO - World Wide Web

JF - World Wide Web

SN - 1386-145X

IS - 3

ER -