Abstract:

Aiming at the dynamic trust evaluation in large-scale distributed systems,which are similar to peer-topeer file-sharing systems,an adaptive dynamic trust evaluation model is constructed based on the genetic algorithm. In this model,the trust relationship between nodes changes dynamically according to their history behaviors and other evaluation data.  Moreover,this model can automatically generate and update evaluation rules with the changes of network environment. Experimental results show that,as compared with the existing trust evaluation models,the proposed model is of a better dynamic adaptive ability in the presence of the dynamic changes of network environment.

Key words: peer to peer networks, dynamic trust evaluation, adaptive ability, genetic algorithm