Research On Microblog Recommendation Based On Tags

As a symbol of Age of Web 2.0, microblog has developed rapidly and has been extensively used in recent years. As a social network platform for information share,spread and acquisition based on user relations, it not only can expand the interpersonal circle and facilitate social contact, but also plays an important medium for people to obtain the latest information and comments from others. How to provide personal services and select high-quality content for users on this platform and how to effectively lower the cost of obtaining useful information for users have become extremely important. Finding users' interests is the premise for designing these personal services.Thus, recommendation algorithm based on user interests should be presented.In the paper, we put forward a user interest model construction method and take the recommendation as research background.Aiming at the characteristics of microblog such as its shortness, limited information, high-dimension and sparsity, tag is considered as a main research object,muti-label correlation and user social relation is integrated to recommend microblog information. The main content is summarized as follows:(1) The conservative approach is taken to extract useful information of microblog,such as word segmentation, splitting on whitespaces and punctuation mark, and eliminating stop words to obtain pure data.(2) A microblog recommendation algorithm based on multi-tag correlation is proposed. On one hand, we develop a user tag retrieval strategy to assign tags for users(those with no tags or only a few tags), and create a user-tag matrix to represent the initial weight of users' tags. On the other hand, we construct a correlation matrix of multi-tag by investigating inner and outer correlation between tags and update the original user-tag matrix to obtain the final weight. Our extensive experimental study shows the scalability and efficiency of this approach.(3) A tag correlation and user social relation based on recommendation approach is presented. On the basis of the previous work, we first investigate the social relation between multi-users and calculate social relation similarity. Then a social relation matrix of multi-user is constructed by analyzing both following and follower information between users and iteratively calculate with the updated user-tag matrix toobtain the final weights.We focus on the three points above on microblog recommendation method.Experiments show the effectiveness of our algorithm.