On the analysis of reputation for agent-based web services

Maintaining sound reputation requires robust control and investigation. In this paper, we analyze a reputation mechanism that objectively maintains accurate reputation evaluation of selfish agent-based web services. In the proposed framework, web services are ranked using their reputation as a result of provided feedback reflecting consumers’ satisfaction. However, selfish web services may alter their public reputation level by managing to get fake feedback. We investigate the payoffs of different scenarios by focusing on the issues that discourage web services to act maliciously. We also analyze the details of the proposed mechanism by discussing simulation and empirical results that fully depict the system parameters and show the feasibility of the proposed approach.

► Theoretical analysis of reputation-based infrastructure for agent-based web services.
► Computation of incentives and penalties to make the system components trustful.
► Extensive simulation study of multiple scenarios confirming the theoretical findings.