Self-Organized Service Management in Social Systems

Humans create efficient social structures in a self-organized way. People tend to join groups with other people with similar characteristics. This is call homophily. This paper proposes how homophily can be introduced in Service-Oriented Multiagent Systems to create efficient self-organized structures in which agents are linked to similar agents, where the similarity is based on the set of services that each agent provides and the roles they play. The results show that a greedy method can be used to locate services in the network and that homophily, which links similar services together, can produce a significant improvement in the performance of the search process. A second contribution is the study of the adaptation of the agents to the number and the type of services demanded. The paper shows how, considering just local information and making local decisions to stay or leave the system, the network adapts itself to a known service distribution.