Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks

Pengfei Li, Hua Lu, Gang Zheng, Qian Zheng, Long Yang, Gang Pan

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

Many e-commerce platforms today allow users to give their rating scores and reviews on items as well as to establish social relationships with other users. As a result, such platforms accumulate heterogeneous data including numeric scores, short textual reviews, and social relationships. However, many recommender systems only consider historical user feedbacks in modeling user preferences. More specifically, most existing recommendation approaches only use rating scores but ignore reviews and social relationships in the user-generated data. In this paper, we propose TSNPF-a latent factor model to effectively capture user preferences and item features. Employing Poisson factorization, TSNPF fully exploits the wealth of information in rating scores, review text and social relationships altogether. It extracts topics of items and users from the review text and makes use of similarities between user pairs with social relationships, which results in a comprehensive understanding of user preferences. Experimental results on real-world datasets demonstrate that our TSNPF approach is highly effective at recommending items to users.
Original languageEnglish
Title of host publicationProceedings of the 30th Web Conference (WWW)
Publication date2019
Pages995-1005
Publication statusPublished - 2019
EventThe Web Conference 2019 (previously WWW) - San Francisco, United States
Duration: 13 May 201917 May 2019
Conference number: 30th
https://www2019.thewebconf.org/

Conference

ConferenceThe Web Conference 2019 (previously WWW)
Number30th
CountryUnited States
CitySan Francisco
Period13/05/201917/05/2019
Internet address

Fingerprint

Recommender systems
Factorization
Feedback

Cite this

Li, P., Lu, H., Zheng, G., Zheng, Q., Yang, L., & Pan, G. (2019). Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks. In Proceedings of the 30th Web Conference (WWW) (pp. 995-1005)
Li, Pengfei ; Lu, Hua ; Zheng, Gang ; Zheng, Qian ; Yang, Long ; Pan, Gang. / Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks. Proceedings of the 30th Web Conference (WWW). 2019. pp. 995-1005
@inproceedings{8e92c9b7f6f242e49689e16554ffe5ce,
title = "Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks",
abstract = "Many e-commerce platforms today allow users to give their rating scores and reviews on items as well as to establish social relationships with other users. As a result, such platforms accumulate heterogeneous data including numeric scores, short textual reviews, and social relationships. However, many recommender systems only consider historical user feedbacks in modeling user preferences. More specifically, most existing recommendation approaches only use rating scores but ignore reviews and social relationships in the user-generated data. In this paper, we propose TSNPF-a latent factor model to effectively capture user preferences and item features. Employing Poisson factorization, TSNPF fully exploits the wealth of information in rating scores, review text and social relationships altogether. It extracts topics of items and users from the review text and makes use of similarities between user pairs with social relationships, which results in a comprehensive understanding of user preferences. Experimental results on real-world datasets demonstrate that our TSNPF approach is highly effective at recommending items to users.",
author = "Pengfei Li and Hua Lu and Gang Zheng and Qian Zheng and Long Yang and Gang Pan",
year = "2019",
language = "English",
pages = "995--1005",
booktitle = "Proceedings of the 30th Web Conference (WWW)",

}

Li, P, Lu, H, Zheng, G, Zheng, Q, Yang, L & Pan, G 2019, Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks. in Proceedings of the 30th Web Conference (WWW). pp. 995-1005, The Web Conference 2019 (previously WWW), San Francisco, United States, 13/05/2019.

Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks. / Li, Pengfei; Lu, Hua; Zheng, Gang; Zheng, Qian; Yang, Long; Pan, Gang.

Proceedings of the 30th Web Conference (WWW). 2019. p. 995-1005.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks

AU - Li, Pengfei

AU - Lu, Hua

AU - Zheng, Gang

AU - Zheng, Qian

AU - Yang, Long

AU - Pan, Gang

PY - 2019

Y1 - 2019

N2 - Many e-commerce platforms today allow users to give their rating scores and reviews on items as well as to establish social relationships with other users. As a result, such platforms accumulate heterogeneous data including numeric scores, short textual reviews, and social relationships. However, many recommender systems only consider historical user feedbacks in modeling user preferences. More specifically, most existing recommendation approaches only use rating scores but ignore reviews and social relationships in the user-generated data. In this paper, we propose TSNPF-a latent factor model to effectively capture user preferences and item features. Employing Poisson factorization, TSNPF fully exploits the wealth of information in rating scores, review text and social relationships altogether. It extracts topics of items and users from the review text and makes use of similarities between user pairs with social relationships, which results in a comprehensive understanding of user preferences. Experimental results on real-world datasets demonstrate that our TSNPF approach is highly effective at recommending items to users.

AB - Many e-commerce platforms today allow users to give their rating scores and reviews on items as well as to establish social relationships with other users. As a result, such platforms accumulate heterogeneous data including numeric scores, short textual reviews, and social relationships. However, many recommender systems only consider historical user feedbacks in modeling user preferences. More specifically, most existing recommendation approaches only use rating scores but ignore reviews and social relationships in the user-generated data. In this paper, we propose TSNPF-a latent factor model to effectively capture user preferences and item features. Employing Poisson factorization, TSNPF fully exploits the wealth of information in rating scores, review text and social relationships altogether. It extracts topics of items and users from the review text and makes use of similarities between user pairs with social relationships, which results in a comprehensive understanding of user preferences. Experimental results on real-world datasets demonstrate that our TSNPF approach is highly effective at recommending items to users.

M3 - Article in proceeding

SP - 995

EP - 1005

BT - Proceedings of the 30th Web Conference (WWW)

ER -

Li P, Lu H, Zheng G, Zheng Q, Yang L, Pan G. Exploiting Ratings, Reviews and Relationships for Item Recommendations in Topic Based Social Networks. In Proceedings of the 30th Web Conference (WWW). 2019. p. 995-1005