Abstract
User information sharing is an important behavior in online social networks. Understanding such behavior could help in various applications such as user modeling, information cascade analysis, viral marketing, etc. In this paper, we aim to understand the strategies users employ to make retweet decision. We are interested in investigating whether these strategies in online social network contain significant information about users and can be used to further characterize users. We propose a flexible model that captures a number of behavior signals affecting user's retweet decision. Our empirical results show that the inferred strategies can help increase the performance of retweet prediction.