A Group-Based Trust Propagation Method

Abstract

Trust is a concept taken from social sciences and is<br><br>considered as a soft security approach that is effective in<br><br>reducing risk. In this paper, for estimating the trust between<br><br>unknown nodes, a group-based trust propagation method has<br><br>been proposed. Most of the conventional trust propagation<br><br>methods are not applicable for trust evaluation of today\\'s large<br><br>trust graphs. Our trust propagation method is scalable because<br><br>of using grouping method. For better trust estimation inside<br><br>group the confidence of trust value have been considered. We<br><br>also consider two factors Intermediate Group Confidence<br><br>(IGC) and Group Confidence (GC) for confidence of trust<br><br>between two groups. To evaluate this method a real large data<br><br>set of advogato.org is used. Evaluation of accuracy is based on<br><br>correlation and mean absolute error(MAE). Comparing the<br><br>proposed method with the Iterative Multiplication method<br><br>(IMS) results suggests that the correlation and absolute mean<br><br>error have been improved. In addition, due to the use of groupbased<br><br>method the speed of the proposed method has been<br><br>improved.