Abstract
The safety of cloud services within a Trust Model System (TMS) is certainly compromised by a lack of defense against security threats as well as by inaccuracy of the trust results. Our proposed model addresses well-known security threats to the reputation trust model system, and is shown to deal with all possible potential attack threats, such as Sybil, on-off, and collusion attacks, by specifying the identity of users and tracking activities undertaken by them in order to easily track unauthorized consumers or attackers and to provide proof of any kind of data leakage. The TMS can also oversee the authorization of whoever uploads feedback into the system. It can also identify invalid feedback and discard it from the system. The algorithms of the TMS first establish a variety of trust criteria in which trustworthiness is calculated. Then, feedback from the cloud service provider nodes is accepted only according to the rules of the TMS. A consumer's trust value is finally computed using a flexible system capable of guaranteeing a good balance of consumer trust and owners' feedback. Furthermore, a majority of the existing TMS models do not take full account of interaction importance, thus impeding the accuracy of the trust values, a shortcoming that has been rectified in our proposed model. (C) 2021 THE AUTHORS. Published by Elsevier BV. on behalf of Faculty of Computers and Artificial Intelligence, Cairo University.