Mining user similarity from semantic trajectories

  • Authors:

  • Josh Jia-Ching Ying;Eric Hsueh-Chan Lu;Wang-Chien Lee;Tz-Chiao Weng;Vincent S. Tseng

  • Affiliations:

  • National Cheng Kung University, Tainan City, Taiwan (R.O.C.);National Cheng Kung University, Tainan City, Taiwan (R.O.C.);Pennsylvania State University, University Park, PA;National Cheng Kung University, Tainan City, Taiwan (R.O.C.);National Cheng Kung University, Tainan City, Taiwan (R.O.C.)

  • Venue:
  • Proceedings of the 2nd ACM SIGSPATIAL International Workshop on Location Based Social Networks
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

In recent years, research on measuring trajectory similarity has attracted a lot of attentions. Most of similarities are defined based on the geographic features of mobile users' trajectories. However, trajectories geographically close may not necessarily be similar because the activities implied by nearby landmarks they pass through may be different. In this paper, we argue that a better similarity measurement should have taken into account the semantics of trajectories. In this paper, we propose a novel approach for recommending potential friends based on users' semantic trajectories for location-based social networks. The core of our proposal is a novel trajectory similarity measurement, namely, Maximal Semantic Trajectory Pattern Similarity (MSTP-Similarity), which measures the semantic similarity between trajectories. Accordingly, we propose a user similarity measurement based on MSTP-Similarity of user trajectories and use it as the basis for recommending potential friends to a user. Through experimental evaluation, the proposed friend recommendation approach is shown to deliver excellent performance.