A new method of trust mirroring estimation based on social networks parameters by fuzzy system

Social networks and semantic web has become more widespread these days. Lots of information has been exchanged between different people daily. The most important issue is the exchange of information between users is trust to the sources and to different users and there is no direct trust because of the vastness of the web in most cases and trust must be predicted in different ways. Extensive researches about trust predicting in social networks, online networks and semantic web has been done in recent years. In this article, we use trust mirroring to predict trust which estimates the amount of trust based on the similarity amount between trustor and trustee. In this article, we use parameters as degree, degree quality and local density in social networks to determine the similarity criteria and we offer fuzzy systems to calculate the amount of similarity for each of these parameters. Then we estimate the amount of trust considering the amount of trustor reputation and amount of similarity. The proposed method provides an acceptable estimation for trust based on the experimental results in the social networks.