A HITS-based POI recommendation algorithm for location-based social networks

  • Authors:
  • Xuelian Long;James Joshi

  • Affiliations:
  • University of Pittsburgh, Pittsburgh, PA;University of Pittsburgh, Pittsburgh, PA

  • Venue:
  • Proceedings of the 2013 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Location-Based Social Networks (LBSNs), (also called as Geo-Social Networks), has been attracting more and more users by providing services that integrate social activities with location information. LBSN systems usually provide support for indicating various Points of Interest (POIs) but there is no straightforward rating mechanism for POIs in most LBSNs [1]. POI recommendations in LBSNs, thus, is an important and challenging research topic. In this paper, we first investigate the dataset crawled from Foursquare to explore the features that attract and influence users to check in at various POIs. Based on the analysis results, we propose a HITS (Hypertext Induced Topic Search)-based POI recommendation algorithm to recommend POIs to LBSN users that can also incorporate the impact of the social relationships on recommendations. We evaluate our proposed model on Foursquare dataset and compare our results with the latest POI recommendation algorithm. The experimental results show that our approach performs better.