Home // International Journal On Advances in Intelligent Systems, volume 9, numbers 3 and 4, 2016 // View article

Computing Similarity between Users on Location-Based Social Networks

Authors:
Soha Mohamed
Alia Abdelmoty

Keywords: GeoFolksonomy; User Profiles; Location-based Social Networks.

Abstract:
With the current trend of embedding location services within social networks, an ever growing amount of users’ spatiotemporal tracks are being collected. These tracks can be used to generate user profiles to reflect users’ interests in places. User-contributed annotations of places, as well as other place properties, add a layer of important semantics that if considered, can result in more refined representations of the users' profiles. In this paper, we study how such place-oriented profiles can be used to represent similarity between users of Location-Based Social Networks (LBSN). Spatial as well as semantic dimensions of the user-provided information are used within a folksonomy data model to represent relationships between users, places and tags. The model allows simple co-occurrence methods and similarity measures to be applied to build different views of personalized user profiles. Basic profiles capture direct user interactions, while enriched profiles offer an extended view of user’s association with places and tags that takes into account relationships in the folksonomy. The main contribution of this work is the demonstration of how the different data dimensions captured on location-based social networks can be combined to represent useful views of user profiles and to compute similarity between users.

Pages: 542 to 553

Copyright: Copyright (c) to authors, 2016. Used with permission.

Publication date: December 31, 2016

Published in: journal

ISSN: 1942-2679