Gradual trust and distrust in recommender systems

  • Authors:
  • Patricia Victor;Chris Cornelis;Martine De Cock;Paulo Pinheiro da Silva

  • Affiliations:
  • Computational Web Intelligence, Department of Applied Mathematics and Computer Science, Ghent University, Krijgslaan 281 (S9), 9000 Gent, Belgium;Computational Web Intelligence, Department of Applied Mathematics and Computer Science, Ghent University, Krijgslaan 281 (S9), 9000 Gent, Belgium;Computational Web Intelligence, Department of Applied Mathematics and Computer Science, Ghent University, Krijgslaan 281 (S9), 9000 Gent, Belgium;Department of Computer Science, The University of Texas at El Paso, Computer Science Building 222B, 500 W University Ave., El Paso, USA

  • Venue:
  • Fuzzy Sets and Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.20

Visualization

Abstract

Trust networks among users of a recommender system (RS) prove beneficial to the quality and amount of the recommendations. Since trust is often a gradual phenomenon, fuzzy relations are the pre-eminent tools for modeling such networks. However, as current trust-enhanced RSs do not work with the notion of distrust, they cannot differentiate unknown users from malicious users, nor represent inconsistency. These are serious drawbacks in large networks where many users are unknown to each other and might provide contradictory information. In this paper, we advocate the use of a trust model in which trust scores are (trust,distrust)-couples, drawn from a bilattice that preserves valuable trust provenance information including gradual trust, distrust, ignorance, and inconsistency. We pay particular attention to deriving trust information through a trusted third party, which becomes especially challenging when also distrust is involved.