Recommender System Based on Random Walk with Topic Model
Weisi Feng, Chenyang Jing, Li Li · 2016
Recommender systems become extremely popular and widely applied in recent years. Researchers have done many work to developing recommender systems in social network. However, recommendation algorithm is still a challenging problem in practice. In this paper, we address the problem of recommending both friends and product simultaneously in the social network. Recommendation system are widely researched in social network with kinds of methods such as Collaborative Filtering. However, most of the methods only recommend relationships or product separately. To address this problem, we propose an User-Item(UI) bipartite graph with Topic Model, which simultaneously incorporates relationships and interest information to model complex relation among users and products. And then we apply Random Walk on the UI bipartite graph to measure the relevance between users and products, and the relevance among users as well. Finally, we evaluate our approach on CiteULike dataset and last.fm dataset. Experiments show the effectiveness of our approach. Comparison with other methods on the two datasets indicates that our approach do make a better job.