A Fuzzy Trust Model for Multi-agent System
In the dynamic and uncertain network environment, the trust is an important mechanism for security and reliability of multi-agent system. This paper proposes a trust model for multi-agent system using fuzzy sets (TMMASFS). The distinguishing feature of TMMASFS is that there is the self-recommendation trust besides the direct trust and the recommendation trust. The self-recommendation trust is useful very much when the manager agent has neither the direct experience nor the recommendation experience about the contractor agents. Then the trust dynamic modification theorem is introduced, which can not only valuate the trust but also monitor the executed process of the task. At last, the trust valuation algorithm is presented, and the result of the experiment shows TMMASFS is efficient and adapted to the dynamic and uncertain network environment.
Unable to display preview. Download preview PDF.
- 1.Ramchurn, S.D., Sierra, C., Godo, L.: A Computational Trust Model for Multi-Agent Interactions based on Confidence and Reputation. In: Proceedings of 6th International Workshop of Deception, Fraud and Trust in Agent Societies, Melbourne, Australia, pp. 69–75 (2003)Google Scholar
- 2.Bower, A.G., et al.: Learning about a Population of Agents and the Evolution of Trust and Cooperation. International Journal of Industrial Organization, 165–190 (1996)Google Scholar
- 3.Xu, F., Lü, J., Zheng, W., Cao, C.: Design of a Trust Valuation Model in Software Service Coordination. Journal of Software, 1043–1051 (2003)Google Scholar
- 4.Gang, C., Ruiqian, L.: The Relation Web Model-An Organizational Approach to Agent Cooperation based on Social Mechanism. Journal of Computer Research and Development, 107–114 (2003)Google Scholar