Automatic trust calculation for service‐oriented systems

Abstract

Abstract-To choose the best service, how much a service can be trusted is increasingly important for service consumers. In addition, manually assigned feedback costs much time and suffers several drawbacks. Automatic trust calculation is the only feasible method for large-scale service-oriented applications. Hence, we propose an automatic trust calculation using nontrust quality criterion values. To make the calculation accurate, we employ the Kalman Filter to filter out malicious non-trust values instead of directly filtering out malicious trust values. Furthermore, to offer higher detection accuracy we propose an improved algorithm based on the joint probability by taking the relationship between the non-trust criterion value and its variance into account. Although malicious data can be filtered out, dishonest or inaccurate values can still influence trust values. Therefore, similarity between consumers is used to weight the values from others, which is calculated by mostly used Euclidean distance function. However, we modify a collection of distance functions to calculate similarity. Finally, experiments are carried out to access the validation and robustness of our model. The results show that our improved algorithm can offer higher detection accuracy under several malicious situations, and we also discovered another similarity function performed better than the Euclidean function.

A Customer-Centric Trust Evaluation Model for Personalized Service Selection

“…A customer's trust in another customer is based on the preference similarity between each other historically. The similarity between consumers is used to weigh ratings submitted by consumers [42]. The higher the preference similarity between and is, the more trusts , and the more trusts .…”