From tie strength to function: Home location estimation in social network

Abstract

In this paper, we focus on the problem of estimating users' home locations in the Twitter network. In order to solve the aforementioned problem, we propose a Social Tie Factor Graph Model (STFGM) for estimating a Twitter user's city-level location based on the following network, user-centric data and tie strength. In STFG, relationships between users and locations in social network are modeled as nodes, the attributes and correlations are modeled as factors. An efficient algorithm is proposed to learn model parameters and to predict unknown relationships. We evaluate our proposed method on large Twitter networks. Experimental results demonstrate that our proposed method significantly outperforms several state-of-the-art methods and achieves the best performance.