Please wait a minute...
Front. Inform. Technol. Electron. Eng.  2013, Vol. 14 Issue (9): 711-721    DOI: 10.1631/jzus.CIIP1303
    
Enhancing recommender systems by incorporating social information
Li-wei Huang, Gui-sheng Chen, Yu-chao Liu, De-yi Li
Institute of Command Information System, PLA University of Science and Technology, Nanjing 210007, China; Institute of Electronic System Engineering, Beijing 100039, China
Enhancing recommender systems by incorporating social information
Li-wei Huang, Gui-sheng Chen, Yu-chao Liu, De-yi Li
Institute of Command Information System, PLA University of Science and Technology, Nanjing 210007, China; Institute of Electronic System Engineering, Beijing 100039, China
 全文: PDF 
摘要: Although recommendation techniques have achieved distinct developments over the decades, the data sparseness problem of the involved user-item matrix still seriously influences the recommendation quality. Most of the existing techniques for recommender systems cannot easily deal with users who have very few ratings. How to combine the increasing amount of different types of social information such as user generated content and social relationships to enhance the prediction precision of the recommender systems remains a huge challenge. In this paper, based on a factor graph model, we formalize the problem in a semi-supervised probabilistic model, which can incorporate different user information, user relationships, and user-item ratings for learning to predict the unknown ratings. We evaluate the method in two different genres of datasets, Douban and Last.fm. Experiments indicate that our method outperforms several state-of-the-art recommendation algorithms. Furthermore, a distributed learning algorithm is developed to scale up the approach to real large datasets.
关键词: Recommender systemSocial informationFactor graph model    
Abstract: Although recommendation techniques have achieved distinct developments over the decades, the data sparseness problem of the involved user-item matrix still seriously influences the recommendation quality. Most of the existing techniques for recommender systems cannot easily deal with users who have very few ratings. How to combine the increasing amount of different types of social information such as user generated content and social relationships to enhance the prediction precision of the recommender systems remains a huge challenge. In this paper, based on a factor graph model, we formalize the problem in a semi-supervised probabilistic model, which can incorporate different user information, user relationships, and user-item ratings for learning to predict the unknown ratings. We evaluate the method in two different genres of datasets, Douban and Last.fm. Experiments indicate that our method outperforms several state-of-the-art recommendation algorithms. Furthermore, a distributed learning algorithm is developed to scale up the approach to real large datasets.
Key words: Recommender system    Social information    Factor graph model
收稿日期: 2013-03-15 出版日期: 2013-09-05
CLC:  TP301.6  
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
Li-wei Huang
Gui-sheng Chen
Yu-chao Liu
De-yi Li

引用本文:

Li-wei Huang, Gui-sheng Chen, Yu-chao Liu, De-yi Li. Enhancing recommender systems by incorporating social information. Front. Inform. Technol. Electron. Eng., 2013, 14(9): 711-721.

链接本文:

http://www.zjujournals.com/xueshu/fitee/CN/10.1631/jzus.CIIP1303        http://www.zjujournals.com/xueshu/fitee/CN/Y2013/V14/I9/711

[1] Imran Ghani, Choon Yeul Lee, Sung Hyun Juhn, Seung Ryul Jeong. Semantics-oriented approach for information interoperability and governance: towards user-centric enterprise architecture management[J]. Front. Inform. Technol. Electron. Eng., 2010, 11(4): 227-240.