融合标签关联关系与用户社交关系的微博推荐方法

Microblog Recommendation Based on Tag Correlation and User Social Relation

通过分析微博特点及现有微博推荐算法的缺陷,提出一种融合了标签间关联关系与用户间社交关系的微博推荐方法.采用标签检索策略对未加标签和标签较少的用户进行加标,构建用户-标签矩阵,得到用户标签权重,为了解决该矩阵中稀疏的问题,通过挖掘标签间的关联关系,继而更新用户-标签矩阵.考虑到多用户之间社交关系对挖掘用户兴趣并进行微博推荐的重要性,构建用户-用户社交关系相似度矩阵,并与更新后的用户-标签矩阵进行迭代,得到最终的用户兴趣并进行相关推荐.实验证明了该算法针对微博信息推荐是有效的.

A novel microblog recommendation method combining the tag correlation with the user social relation is proposed via analyzing microblog features and the deficiencies of existing microblog recommendation algorithm. Specifically,we establish a tag retrieval strategy to add tags for unlabeled users and users with fewtags,and then build the user-tag matrix and obtain user-tag weights. In order to solve the problem of sparsity of the matrix,we investigate the correlation between the tags to update the user-tag matrix. Considering the significance of user social relation for microblog recommendation,a user-user social relation similarity matrix is constructed and a mechanism is designed to iteratively obtain user interest. Experimental results showthat the algorithm is effective in microblog recommendation.