¹State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing Wuhan University, Wuhan 430079, China
²College of Computer Science, South-Central University for Nationalities, Wuhan, Hubei, China, 430074

---

Abstract—Social relationships among Agents are natural, since Agents interact with each other when executing tasks. Diversity of the types and strengths of the social relationships influences the system policy of resource application and how Agents interact and cooperate with peers. AgentTMS is proposed by leveraging the Agent social relationships to improve the traditional trust models based on Agent’s reputation and activity. Three types of Agent social relationships basing on biological relations, common social ties and business activity are defined and featured. The strength of the social relationships are measured and integrated into AgentTMS. The performance of AgentTMS is evaluated and validated by the presented results through considering the faster approach to the optimal resource allocation strategy. More detail here the initialization to the individual Agent and the effect to the rate of system reaching efficient state are about the result descriptions.

Index Terms—MAS, Trust Model, Social Relationship.

[PDF]

Cite: Min Peng, ZhengQuan Xu, ShaoMing Pan, Rui Li, Tengyue Mao, "AgentTMS: A MAS Trust Model based on Agent Social Relationship," Journal of Computers vol. 7, no. 6, pp. 1535-1542, 2012.