The application of collaborative filtering for trust management in p2p communities

  • Authors:
  • Min Zuo;Kai Wang;Jianhua Li

  • Affiliations:
  • Department of Electronic Engineering, Shanghai Jiaotong University, Shanghai, China;Department of Electronic Engineering, Shanghai Jiaotong University, Shanghai, China;Department of Electronic Engineering, Shanghai Jiaotong University, Shanghai, China

  • Venue:
  • ISPA'05 Proceedings of the Third international conference on Parallel and Distributed Processing and Applications
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

The open and anonymous nature of P2P services opens the door to malicious peers who cause the loss of trust by providing corrupted data or harmful services. The introduction of a trust management system is one of the possible ways to combat this problem. This paper presents some new ideas for the design of a P2P trust management system. Its main contributions include: a recommendation-aggregating model based on collaborative filtering (CF), a polling protocol for trust queries and responses, and the use of identity-based cryptosystem to secure recommendations. Simulations show that our CF-based trust model performs pretty well even when malicious peers make the majority.