Interest propagation based recommendations in social networks: A case study

摘要

In this paper, we propose an interest propagation based method, which incorporates the interest propagation over weighted social relationships for top-k item recommendation in social networks. Our method combines the user-item click information and user social action information for recommendations. We study the effectiveness of our method on Sina Weibo, one of the most popular social network sites in China. The experiments are conducted to evaluate the performance of our methods and results show that our method outperforms the traditional collaborative filtering based method.