Detecting occasional reputation attacks on cloud services

Abstract

Cloud service consumers' feedback is a good source to assess the trustworthiness of cloud services. However, it is not unusual that a trust management system experiences malicious behaviors from its users. Although several techniques have been proposed to address trust management in cloud environments, the issue of how to detect occasional reputation attacks on cloud services is still largely overlooked. In this paper, we introduce an occasional attacks detection model that recognizes misleading trust feedbacks from occasional collusion and Sybil attacks and adjusts trust results for cloud services that have been affected by these malicious behaviors. We have collected a large collection of consumer's trust feedbacks given on real-world cloud services (over ten thousand records) to evaluate and demonstrate the applicability of our approach and show the capability of detecting such malicious behaviors. © 2013 Springer-Verlag Berlin Heidelberg.