Identifying Users Based on Behavioral-Modeling across Social Media Sites

Social media is playing an important role in our daily life. People usually hold various identities on different social media sites. User-contributed Web data contains diverse information which reflects individual interests, political opinions and other behaviors. To integrate these behaviors information, it is of value to identify users across social media sites. This paper focuses on the challenge of identifying unknown users across different social media sites. A method to relate user’s identities across social media sites by mining users’ behavior information and features is introduced. The method has two key components. The first component distinguishes different users by analyzing their common social network behaviors and finding strong opposing characters. The second component constructs a model of behavior features that helps to obtain the difference of users across social media sites. The method is evaluated through two experiments on Twitter and Sina Weibo. The results of experiments show that the method is effective.