Research on selective trust-path search and aggregation in distributed environment

Abstract

Current trust models based on reputation had some disadvantages in characterizing recommendation credibility and dealing with Trust-Path Search and aggregation.To solve these problems,an updating algorithm for recommendation credibility was proposed considering factors of time decay,recommendation inosculation and success transaction rate.Furthermore,a selective search-algorithm for trust-path in distributed environment was presented.The algorithm uses recommendation credibility,evaluation similarity and trust-path length as its control parameters and can evade malicious nodes directly in the process of searching.It can also stop the search for unnecessary trust-paths and execute the search for paths containing valuable recommendation.Lastly,trust aggregation method was given by using an advanced combining algorithm in D-S theory of evidence.Simulation results show that compared with existing trust models,the proposed model is more robust on defending attacks of various malicious nodes.

Keywords

• distributed network;trust model;recommendation credibility;recommendation trust path;D-S theory of evidence