Learning dual preferences with non-negative Matrix tri-factorization  for Top-N recommender system

Xiangsheng LI; Yanghui RAO; Haoran XIE; Yufu CHEN; Raymond Y. K. LAU; Fu Lee WANG; Jian YIN

doi:10.1007/978-3-319-91452-7_9

Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system

Xiangsheng LI, Yanghui RAO^*, Haoran XIE, Yufu CHEN, Raymond Y. K. LAU, Fu Lee WANG, Jian YIN

^*Corresponding author for this work

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research

1 Citation (Scopus)

Abstract

In recommender systems, personal characteristic is possessed by not only users but also displaying products. Users have their personal rating patterns while products have different characteristics that attract users. This information can be explicitly exploited from the review text. However, most existing methods only model the review text as a topic preference of products, without considering the perspectives of users and products simultaneously. In this paper, we propose a user-product topic model to capture both user preferences and attractive characteristics of products. Different from conventional collaborative filtering in conjunction with topic models, we use non-negative matrix tri-factorization to jointly reveal the characteristic of users and products. Experiments on two real-world data sets validate the effectiveness of our method in Top-N recommendations.

Original language	English
Title of host publication	Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I
Editors	Jian PEI, Yannis MANOLOPOULOS, Shazia SADIQ, Jianxin LI
Publisher	Springer International Publishing AG
Pages	133-149
Number of pages	17
ISBN (Electronic)	9783319914527
ISBN (Print)	9783319914510
DOIs	https://doi.org/10.1007/978-3-319-91452-7_9
Publication status	Published - 2018
Externally published	Yes
Event	23rd International Conference on Database Systems for Advanced Applications - Gold Coast, Australia Duration: 21 May 2018 → 24 May 2018 http://www.ict.griffith.edu.au/conferences/dasfaa2018/

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	10827
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	23rd International Conference on Database Systems for Advanced Applications
Abbreviated title	DASFAA 2018
Country/Territory	Australia
Period	21/05/18 → 24/05/18
Internet address	http://www.ict.griffith.edu.au/conferences/dasfaa2018/

Bibliographical note

We are grateful to the anonymous reviewers for their valuable comments on this manuscript. The research has been supported by the National Natural Science Foundation of China (61502545, U1611264, U1711262), a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (UGC/FDS11/E03/16), and the Individual Research Scheme of the Dean’s Research Fund 2017–2018 (FLASS/DRF/IRS-8) of The Education University of Hong Kong.

Keywords

Top-N recommender system
Topic model
Matrix tri-factorization

Access to Document

10.1007/978-3-319-91452-7_9

Cite this

LI, X., RAO, Y., XIE, H., CHEN, Y., LAU, R. Y. K., WANG, F. L., & YIN, J. (2018). Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system. In J. PEI, Y. MANOLOPOULOS, S. SADIQ, & J. LI (Eds.), Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I (pp. 133-149). (Lecture Notes in Computer Science; Vol. 10827). Springer International Publishing AG. https://doi.org/10.1007/978-3-319-91452-7_9

LI, Xiangsheng ; RAO, Yanghui ; XIE, Haoran et al. / Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system. Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I. editor / Jian PEI ; Yannis MANOLOPOULOS ; Shazia SADIQ ; Jianxin LI. Springer International Publishing AG, 2018. pp. 133-149 (Lecture Notes in Computer Science).

@inproceedings{59be1ca935864531959b1b07e00b8553,

title = "Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system",

abstract = "In recommender systems, personal characteristic is possessed by not only users but also displaying products. Users have their personal rating patterns while products have different characteristics that attract users. This information can be explicitly exploited from the review text. However, most existing methods only model the review text as a topic preference of products, without considering the perspectives of users and products simultaneously. In this paper, we propose a user-product topic model to capture both user preferences and attractive characteristics of products. Different from conventional collaborative filtering in conjunction with topic models, we use non-negative matrix tri-factorization to jointly reveal the characteristic of users and products. Experiments on two real-world data sets validate the effectiveness of our method in Top-N recommendations.",

keywords = "Top-N recommender system, Topic model, Matrix tri-factorization",

author = "Xiangsheng LI and Yanghui RAO and Haoran XIE and Yufu CHEN and LAU, {Raymond Y. K.} and WANG, {Fu Lee} and Jian YIN",

note = "We are grateful to the anonymous reviewers for their valuable comments on this manuscript. The research has been supported by the National Natural Science Foundation of China (61502545, U1611264, U1711262), a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (UGC/FDS11/E03/16), and the Individual Research Scheme of the Dean{\textquoteright}s Research Fund 2017–2018 (FLASS/DRF/IRS-8) of The Education University of Hong Kong.; 23rd International Conference on Database Systems for Advanced Applications, DASFAA 2018 ; Conference date: 21-05-2018 Through 24-05-2018",

year = "2018",

doi = "10.1007/978-3-319-91452-7_9",

language = "English",

isbn = "9783319914510",

series = "Lecture Notes in Computer Science",

publisher = "Springer International Publishing AG",

pages = "133--149",

editor = "Jian PEI and Yannis MANOLOPOULOS and Shazia SADIQ and Jianxin LI",

booktitle = "Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I",

address = "Switzerland",

url = "http://www.ict.griffith.edu.au/conferences/dasfaa2018/",

}

LI, X, RAO, Y, XIE, H, CHEN, Y, LAU, RYK, WANG, FL & YIN, J 2018, Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system. in J PEI, Y MANOLOPOULOS, S SADIQ & J LI (eds), Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I. Lecture Notes in Computer Science, vol. 10827, Springer International Publishing AG, pp. 133-149, 23rd International Conference on Database Systems for Advanced Applications, Queensland, Australia, 21/05/18. https://doi.org/10.1007/978-3-319-91452-7_9

Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system. / LI, Xiangsheng; RAO, Yanghui; XIE, Haoran et al.
Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I. ed. / Jian PEI; Yannis MANOLOPOULOS; Shazia SADIQ; Jianxin LI. Springer International Publishing AG, 2018. p. 133-149 (Lecture Notes in Computer Science; Vol. 10827).

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research

TY - GEN

T1 - Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system

AU - LI, Xiangsheng

AU - RAO, Yanghui

AU - XIE, Haoran

AU - CHEN, Yufu

AU - LAU, Raymond Y. K.

AU - WANG, Fu Lee

AU - YIN, Jian

N1 - We are grateful to the anonymous reviewers for their valuable comments on this manuscript. The research has been supported by the National Natural Science Foundation of China (61502545, U1611264, U1711262), a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (UGC/FDS11/E03/16), and the Individual Research Scheme of the Dean’s Research Fund 2017–2018 (FLASS/DRF/IRS-8) of The Education University of Hong Kong.

PY - 2018

Y1 - 2018

N2 - In recommender systems, personal characteristic is possessed by not only users but also displaying products. Users have their personal rating patterns while products have different characteristics that attract users. This information can be explicitly exploited from the review text. However, most existing methods only model the review text as a topic preference of products, without considering the perspectives of users and products simultaneously. In this paper, we propose a user-product topic model to capture both user preferences and attractive characteristics of products. Different from conventional collaborative filtering in conjunction with topic models, we use non-negative matrix tri-factorization to jointly reveal the characteristic of users and products. Experiments on two real-world data sets validate the effectiveness of our method in Top-N recommendations.

AB - In recommender systems, personal characteristic is possessed by not only users but also displaying products. Users have their personal rating patterns while products have different characteristics that attract users. This information can be explicitly exploited from the review text. However, most existing methods only model the review text as a topic preference of products, without considering the perspectives of users and products simultaneously. In this paper, we propose a user-product topic model to capture both user preferences and attractive characteristics of products. Different from conventional collaborative filtering in conjunction with topic models, we use non-negative matrix tri-factorization to jointly reveal the characteristic of users and products. Experiments on two real-world data sets validate the effectiveness of our method in Top-N recommendations.

KW - Top-N recommender system

KW - Topic model

KW - Matrix tri-factorization

UR - http://www.scopus.com/inward/record.url?scp=85048033422&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-91452-7_9

DO - 10.1007/978-3-319-91452-7_9

M3 - Conference paper (refereed)

SN - 9783319914510

T3 - Lecture Notes in Computer Science

SP - 133

EP - 149

BT - Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I

A2 - PEI, Jian

A2 - MANOLOPOULOS, Yannis

A2 - SADIQ, Shazia

A2 - LI, Jianxin

PB - Springer International Publishing AG

T2 - 23rd International Conference on Database Systems for Advanced Applications

Y2 - 21 May 2018 through 24 May 2018

ER -

LI X, RAO Y, XIE H, CHEN Y, LAU RYK, WANG FL et al. Learning dual preferences with non-negative Matrix tri-factorization for Top-N recommender system. In PEI J, MANOLOPOULOS Y, SADIQ S, LI J, editors, Database Systems for Advanced Applications : 23rd International Conference, DASFAA 2018, Gold Coast, QLD, Australia, May 21-24, 2018, proceedings, part I. Springer International Publishing AG. 2018. p. 133-149. (Lecture Notes in Computer Science). Epub 2018 May 13. doi: 10.1007/978-3-319-91452-7_9