Collaborative Model for Predicting Retweeting Behaviors on Twitter

摘要

Nowadays, Twitter has become one of the most important ways for information sharing. Users can spread information they like by retweeting. However, with the growth of twitter, users are easily overwhelmed by large amount of data and it is very diffcult for users to dig out information that they are interested in. To address this problem, we predict tweets that users are really interested in and help them reduce the effort to find useful information. In this paper, we introduce the users' similarity and trust based on retweeting behaviors and propose a retweeting behaviors prediction model based on collaborative filtering. The experiments show that our model was applicable on the real-life data.