基于标签的大众标注系统协同推荐算法  

A Collaborative Recommendation Algorithm in Social Annotation System Based on Tag

作　　者：蒋翠清[1,2] 张玉[1] 陆文星[1] 丁勇[1] 

机构地区：[1]合肥工业大学管理学院,合肥230009 [2]过程优化与智能决策教育部重点实验室,合肥230009

出　　处：《情报学报》2011年第11期1152-1157,共6页Journal of the China Society for Scientific and Technical Information

基　　金：国家自然科学基金项目（70871034,70771037）;安徽省高校自然科学研究重点项目（KJ2010A259）;教育部人文社会科学基金项目（09YJC630055）.

摘　　要：本文针对大众标注系统中现有基于标签的推荐算法的不足，分析了大众标注系统中用户标注的潜在语义。提出了基于标签的大众标注系统协同推荐算法。新的算法利用扩展的PLSA模型将用户标注映射到具有明确意义的语义主题上，较好地消除了标签的语义模糊问题，提高了推荐精度。最后通过实验证明了本文提出的推荐算法效果要优于传统的推荐算法。In allusion to the insufficiency of existing recommendation algorithm based on tags in the social annotation system, this paper analyses the latent semantics of user annotation and proposes a new collaborative recommendation algorithm based on tags in social annotation system. The new algorithm eliminates the semantic ambiguity problem of user annotation by mapping the annotation to well-defined semantic topics using extended PLSA model. Thereby, the precision of the new algorithm is improved. Lastly, the experiment proves the proposed algorithm is better than the traditional algorithm.

关 键 词：大众标注 潜在语义 PLSA 推荐 

分 类 号：G250.76[文化科学—图书馆学]