Exploiting location-aware social networks for efficient spatial query processing

Liang Tang, Haiquan Chen, Wei Shinn Ku, Min Te Sun

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

In this paper, we introduce two watchtower-based parameter-tunable frameworks for efficient spatial processing with sparse distributions of Points of Interest (POIs) by exploiting mobile users’ check-in data collected from the location-aware social networks. In our proposed frameworks, the network traversal can terminate earlier by retrieving the distance information stored in watchtowers. More important, by observing that people’s movement often exhibits a strong spatial pattern, we employ Bayesian Information Criterion-based cluster analysis to model mobile users’ check-in data as a mixture of 2-dimensional Gaussian distributions, where each cluster corresponds to a geographical hot zone. Afterwards, POI watchtowers are established in the hot zones and non-hot zones discriminatorily. Moreover, we discuss the optimal watchtower deployment mechanism in order to achieve a desired balance between the off-line pre-computation cost and the on-line query efficiency. Finally, the superiority of our solutions over the state-of-the-art approaches is demonstrated using the real data collected from Gowalla with large-scale road networks.

Original languageEnglish
Pages (from-to)33-55
Number of pages23
JournalGeoInformatica
Volume21
Issue number1
DOIs
StatePublished - 1 Jan 2017

Keywords

  • Advanced traveler information systems
  • Location-based services
  • Query processing

Fingerprint

Dive into the research topics of 'Exploiting location-aware social networks for efficient spatial query processing'. Together they form a unique fingerprint.

Cite this