Semantic-Based Web Service Discovery Algorithm

In recent years, Web services as a new model of Web applications and a new distributed computing model has gradually become an effective mechanism of integrating data and information, and developing very rapidly. The purpose of Web services is to resolve ingtegration and sharing of data and applications on the heterogeneous platforms. But with the increasingly widespread application of Web services and service requesters often have to face a large number of forms of Web services, how to accurately and efficiently find the service is undoubtedly very important. Complete model of Web services include: service description, service registry, service matching. However, traditional Web service discovery has the obvious disadvantages: service description does not provide services functional semantic description; service registry can not understand the semantic information of Web Services; service matching is based on keywords; Single Web Service may only call the function to complete only a single function, whereas the organic combination of multiple Web Services will be able to complete a series of complex tasks, If single Web Service exchanges interactive models, then the service request client must be updated by programmer to deal with changes, it is not flexible enough. In this context, Semanatic Web Services become a new direction. Semantic Web services combines the advantages of traditional Web Services technology and Semantic Web technology, decribing the functional properties, non-functional properties and behavior of Web Services by using semantic information, providing an effective support for Web Service discovery, execution and automation of service combination. The W3C's Web inventor Tim Berners-Lee announced: "W3C Semantic Web services is one of three major research themes."In the view of current problems of Web Service Discovery, in this paper the ontology-based cluster partition approach for computing the semantic distance between concepts is proposed, which using semantic distance of two concepts compute the concept similarity, and then using multi-level semantic Web Service matching tactic to compute the similarity of servives. For this reason, the following three parts are studied in this paper:Semantic Distance ComputeIn this paper the ontology-based cluster partition approach for computing the semantic distance between concepts is proposed. and then the approach for computing the semantic distance between concepts within single cluster as well as cross-cluster is put forward. In the proposed approach, the non-symmetry of semantic similarities in the pairs of hyponymy concepts is worked out by introducing the forward semantic distance and the reverse semantic distance, the base relation in ontology include equavalation relation, SubClassOf relation, self-define relation belong to the other binary relationships, and the other binary relationships of the pairs of non-hyponymy concepts are deal with by dynamically allocating the relation weights in the light of the locations of concept nodes. On this basis, the calculation of semantic distance.The computing of semantic similarityThe purpose of semantic distance is for the compute of semantic similartiy, in this paper, semantic similarity include three kinds of similarity:concept similarity, concept set similarity, service similarity. This paper also show how to compute this similarity.Algorithm of Semantic Web Service MatchBased on the I/O concept semantic similarity, this paper proposes a two-level matching tactic for Semantic Web Service, First, OWL-S is extended to support the service interface dependencies stated. It is allowed for users to set up dependency weights from output parameters to input parameters according to their own needs in their service requests, and matching weights of input interface and output interface in the calculation of service matching as well. Based on the above, a two-level matching strategy for semantic Web services is proposed. The strategy integrates a single advertising service-based matching algorithm with an advertising service combination-based matching one. The former applies to the case of one service request matching one advertising service. First of all, the later breaks a complex service request down into a simple, single output and multi-input sequence of sub-service requests, then the service matching is popularized to such a situation where a complex service request asks multiple advertising services to complete its task in the collaborative way by exploiting the mechanism for sub-service requests matching advertising service composition chains.This paper devoloped a Semantic-Based Web Services prototype system, and designed every functional model The Query Proeesser supply the user interface for users to prescribe the query condition, and change the Services request to normative request profile according to OWL-S profile. OWL-S/UDDI Converter establishes the mapping between the OWL-S Profile and UDDI to enlarge the UDDI so as to storage the semantic information. Semantic service matching engineer is designed for achieving the service functional-based semantic match. The Service Discovery and Matching Model is the focus of this paper. Finally, using the BookMarket Ontology and Web Services of Buying books as example to analyze the capability of this model.