Integrating Social Relations into Personalized Tag Recommendation
Personalized tag recommendation is to provide a user with a ranked list of tags for a specific resource that best serves the users needs. In this paper, we proposed a personalized tag recommendation algorithm incorporating with users social relations. We model the social annotations made by the collaborative users and the soda) relations between them with a graph model. We associate each node in this graph with a tag preference vector, which is then refined through a random walk procedure over this graph. The tag preferences of the active user and resource are finally combined to generate the recommended tags. We conduct experiments on the Delicious. Experimental results demonstrate the effectiveness of the proposed algorithm.
social tagging tag recommendation personalization
Kaipeng Liu Binxing Fang
Harbin Institute of Technology,Harbin,China Institute of Computing Technology,Chinese Academy of Sciences,Beijing,China
国际会议
南京
英文
292-295
2010-08-26（万方平台首次上网日期，不代表论文的发表时间）