Exploring Trust Propagation Behaviors in Online Communities

摘要

Trust plays a very important role in people's real lives. As online social networking becomes more and more popular nowadays, it is urgent to use trust in online communities, like Amazon.com, Epinions.com, Facebook, Twitter and so on. Unfortunately, trust links in online communities are sparse compared with the number of pairs of users. To address this problem, computer scientists are now using trust propagation to infer trust links using existing information. However, users show different behavior patterns in different online communities. There are two classic theories in sociology, balance theory and status theory, aiming to cope with this problem. In this paper, we use our trust management systems to analyze these two social theories in two real datasets: Epinions.com and Twitter.