Mobile services discovery and selection in the publish/subscribe paradigm

  • Authors:
  • A. M. Roumani;D. B. Skillicorn

  • Affiliations:
  • School of Computing, Queen's University, Kingston, Canada;School of Computing, Queen's University, Kingston, Canada

  • Venue:
  • CASCON '04 Proceedings of the 2004 conference of the Centre for Advanced Studies on Collaborative research
  • Year:
  • 2004

Quantified Score

Hi-index 0.01

Visualization

Abstract

In a publish/subscribe paradigm, user service discovery requires matching user preferences to available published services, e.g., a user may want to find if there is a Chinese restaurant close by. This is a difficult problem when users are mobile, wirelessly connected to a network, and dynamically roaming in different environments. The magnitude of the problem increases with respect to the number of attributes for each users' preference criteria, as matches must be done in real-time. We present an algorithm that uses Singular Value Decomposition to encode each service properties in a few values. Users' preference criteria are matched by using the same encoding to produce a value that can be rapidly compared to those of the services. We show that reasonable matches can be found in time O(m log n) where n is the number of publications and m the number of attributes in the preference criteria (subscription). This is in contrast to 'approximate nearest neighbor' techniques, which require either time or storage exponential in m.