CFUI: Collaborative filtering with unlabeled items

Jing Peng, Dajun Zeng, Bing Liu, Huimin Zhao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

As opposed to Web search, social tagging can be considered an alternative technique tapping into the wisdom of the crowd for organizing and discovering information on the Web. Effective tag-based recommendation of information items is critical to the success of this social information discovery mechanism. Over the past few years, there have been a growing number of studies aiming at improving the item recommendation quality of collaborative filtering (CF) methods by leveraging tagging information. However, a critical problem that often severely undermines the performance of tag-based CF methods, i.e., sparsity of user-item and user-tag interactions, is still yet to be adequately addressed. In this paper, we propose a novel learning framework, which deals with this data sparsity problem by making effective use of unlabeled items and propagating users' preference information between the item space and the tag space. Empirical evaluation using real-world tagging data demonstrates the utility of the proposed framework.

Original languageEnglish (US)
Title of host publicationProceedings of 20th Annual Workshop on Information Technologies and Systems
PublisherSocial Science Research Network
StatePublished - 2010
Event20th Annual Workshop on Information Technologies and Systems, WITS 2010 - St. Louis, MO, United States
Duration: Dec 11 2010Dec 12 2010

Other

Other20th Annual Workshop on Information Technologies and Systems, WITS 2010
CountryUnited States
CitySt. Louis, MO
Period12/11/1012/12/10

Fingerprint

Collaborative filtering

Keywords

  • Social tagging
  • Sparsity
  • Tag-based recommendation
  • Unlabeled items

ASJC Scopus subject areas

  • Information Systems

Cite this

Peng, J., Zeng, D., Liu, B., & Zhao, H. (2010). CFUI: Collaborative filtering with unlabeled items. In Proceedings of 20th Annual Workshop on Information Technologies and Systems Social Science Research Network.

CFUI : Collaborative filtering with unlabeled items. / Peng, Jing; Zeng, Dajun; Liu, Bing; Zhao, Huimin.

Proceedings of 20th Annual Workshop on Information Technologies and Systems. Social Science Research Network, 2010.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Peng, J, Zeng, D, Liu, B & Zhao, H 2010, CFUI: Collaborative filtering with unlabeled items. in Proceedings of 20th Annual Workshop on Information Technologies and Systems. Social Science Research Network, 20th Annual Workshop on Information Technologies and Systems, WITS 2010, St. Louis, MO, United States, 12/11/10.
Peng J, Zeng D, Liu B, Zhao H. CFUI: Collaborative filtering with unlabeled items. In Proceedings of 20th Annual Workshop on Information Technologies and Systems. Social Science Research Network. 2010
Peng, Jing ; Zeng, Dajun ; Liu, Bing ; Zhao, Huimin. / CFUI : Collaborative filtering with unlabeled items. Proceedings of 20th Annual Workshop on Information Technologies and Systems. Social Science Research Network, 2010.
@inproceedings{58be75cbd1e94f2ea57fd83491567149,
title = "CFUI: Collaborative filtering with unlabeled items",
abstract = "As opposed to Web search, social tagging can be considered an alternative technique tapping into the wisdom of the crowd for organizing and discovering information on the Web. Effective tag-based recommendation of information items is critical to the success of this social information discovery mechanism. Over the past few years, there have been a growing number of studies aiming at improving the item recommendation quality of collaborative filtering (CF) methods by leveraging tagging information. However, a critical problem that often severely undermines the performance of tag-based CF methods, i.e., sparsity of user-item and user-tag interactions, is still yet to be adequately addressed. In this paper, we propose a novel learning framework, which deals with this data sparsity problem by making effective use of unlabeled items and propagating users' preference information between the item space and the tag space. Empirical evaluation using real-world tagging data demonstrates the utility of the proposed framework.",
keywords = "Social tagging, Sparsity, Tag-based recommendation, Unlabeled items",
author = "Jing Peng and Dajun Zeng and Bing Liu and Huimin Zhao",
year = "2010",
language = "English (US)",
booktitle = "Proceedings of 20th Annual Workshop on Information Technologies and Systems",
publisher = "Social Science Research Network",

}

TY - GEN

T1 - CFUI

T2 - Collaborative filtering with unlabeled items

AU - Peng, Jing

AU - Zeng, Dajun

AU - Liu, Bing

AU - Zhao, Huimin

PY - 2010

Y1 - 2010

N2 - As opposed to Web search, social tagging can be considered an alternative technique tapping into the wisdom of the crowd for organizing and discovering information on the Web. Effective tag-based recommendation of information items is critical to the success of this social information discovery mechanism. Over the past few years, there have been a growing number of studies aiming at improving the item recommendation quality of collaborative filtering (CF) methods by leveraging tagging information. However, a critical problem that often severely undermines the performance of tag-based CF methods, i.e., sparsity of user-item and user-tag interactions, is still yet to be adequately addressed. In this paper, we propose a novel learning framework, which deals with this data sparsity problem by making effective use of unlabeled items and propagating users' preference information between the item space and the tag space. Empirical evaluation using real-world tagging data demonstrates the utility of the proposed framework.

AB - As opposed to Web search, social tagging can be considered an alternative technique tapping into the wisdom of the crowd for organizing and discovering information on the Web. Effective tag-based recommendation of information items is critical to the success of this social information discovery mechanism. Over the past few years, there have been a growing number of studies aiming at improving the item recommendation quality of collaborative filtering (CF) methods by leveraging tagging information. However, a critical problem that often severely undermines the performance of tag-based CF methods, i.e., sparsity of user-item and user-tag interactions, is still yet to be adequately addressed. In this paper, we propose a novel learning framework, which deals with this data sparsity problem by making effective use of unlabeled items and propagating users' preference information between the item space and the tag space. Empirical evaluation using real-world tagging data demonstrates the utility of the proposed framework.

KW - Social tagging

KW - Sparsity

KW - Tag-based recommendation

KW - Unlabeled items

UR - http://www.scopus.com/inward/record.url?scp=84900435860&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84900435860&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84900435860

BT - Proceedings of 20th Annual Workshop on Information Technologies and Systems

PB - Social Science Research Network

ER -