Agent Service Matchmaking Based on Description Logic

Service matchmaking is one of the major problems in the area of intelligent agent and multi-agent systems. Description logic is the formal tool of knowledge representation and reasoning. Using effective reasoning capability of description logic, especially its effective judgment to concept subsumption, the combination of description logic and service reasoning in multi-agent systems is studied. Adequately making use of effective model-theoretic semantics and concept hierarchy of description logic, the description logic based service matchmaking algorithm for multi-agent systems is proposed, and the automatic service hierarchy and service matchmaking by description logic theory and reasoning mechanism are studied in detail, and five kinds of algorithms of agent service matchmaking, i.e. agent service matchmaking algorithm, strong service matchmaking algorithm, k-nearest service matchmaking algorithm, k-approximate service matchmaking algorithm, m-nearest and n-approximate service matchmaking algorithm, are proposed. These matchmaking methods are based on semantic service matchmaking. Through the mechanism of the service hierarchy, an effective and efficient agent service matchmaking in multi-agent systems is implemented. Because the service matchmaking algorithms based on semantic distance need a lot of hard work by hand and the semantic distance is subjective and uncertain, so the service matchmaking algorithms in this paper overcome the shortcoming of the method based on semantic distance. At last, the realization method of the five kinds of algorithms of agent service matchmaking in Multi-AGent Environment MAGE is introduced.