A spatial-temporal analysis of users' geographical patterns in social media: A case study on microblogs

Abstract

With the development of information technologies, Social Media platforms have become popular and accumulated numerous data about individuals' behavior. It offers a promising opportunity of discovering usable knowledge about the individuals' movement behavior, which fosters novel applications and services. In this paper, in order to study the relations between communities and location clusters, we propose the index of location entropy to measure the degree of dispersion of the locations in each community, and the index of community entropy to measure the degree of dispersion of the communities in each location cluster. At last, we analyze users' trajectories and define four Trajectory Patterns. An algorithm is also proposed to extract those patterns from microblog data. We implement the algorithm and find some interesting and useful results for the intelligent recommender systems. © 2014 Springer-Verlag Berlin Heidelberg.