An intermediary utility-based service search and structure organization approach in service-oriented MAS

Abstract

In distributed service-oriented multiagent systems (SOMASs), agents cooperate autonomously to accomplish service discovery. In a SOMAS application, each agent plays two roles: (i) satisfying the service demands as a service provider, and (ii) forwarding the received queries as an intermediary agent. This study defines that the intermediary utility of an agent is positively related to the success rate of the service discovery in which it participates. If the intermediary utility of an agent is high (e.g., when the neighbors of the agent have abundant service resources), then the agent can better guide service searches. Based on this characteristic, this study proposes an intermediary utilities-based service search and structure organization approach to improve the efficiency of service discovery. In this approach, a service query has a higher probability of being forwarded to agents with higher intermediary utilities, which accelerates the service search. Additionally, the connectivity of the system structure can be preserved by the proposed approach, which ensures that the services can be accessed globally in the system. Extensive experiments demonstrate that the proposed approach can achieve better performance regarding the success rate of service discovery and average forwarding time of service queries comparing to previous approaches.