Evolving service semantics cooperatively: a consumer-driven approach

Abstract

Commerce relies on dynamic creation and modification of services. New service offerings or service demands come into play frequently. Whereas traditional commerce supports creation of new service demands from consumers, e-commerce has so far expected service providers to come up with desirable new service offerings and assigned service consumers a passive role in the process. That is, current e-commerce architecture lacks a consumer-driven approach for the generation of new service descriptions. This paper bridges this gap by proposing a multiagent system of consumers that represent their service needs semantically using ontologies. Using our proposed approach, agents can create new service descriptions, share them with interested others, and use service descriptions that are created by other agents. Hence, more accurate concepts describing consumers' service needs are cooperatively and iteratively created. This leads to a society of consumers with different but overlapping ontologies where mutually accepted service concepts emerge based on consumers' exchange of service descriptions. Our simulations of consumer societies show that allowing cooperative evolution of service ontologies facilitates better representation of consumers' service needs. Further, through cooperation, not only more useful service concepts emerge over time, but also ontologies of consumers having similar service needs become aligned gradually.