Information filtering ia trust relationships diffusion process
Ling Jiao Chen, Lei Gao, Hui Qiong Yang · 2016
Recommender systems benefit us in tackling the problem of information overload and finding potential objects that we are interested in among diverse objects. A variety of recommendation algorithms have been proposed. Most of them only focus on the relationship between users and objects, but neglect the influence of social relationships. In this paper, by considering users' social trust relationships, we propose a trust-based information filtering algorithm with two tunable parameters. Interestingly, experiments on two data sets show the universal optimal parameters for our method. In contrast to original algorithm, our method improve personalized recommendations performance evidently.