Research On Discovery Of Semantic Web Service Based On P2P

With the rapid development of Internet technology, a large number of Web services are deployed in different operation platforms. How to discover and call these services is considered as hotspots of research by the academics and business.Nowadays, the well-known centralized Web service discovery mechanism in business has some problems such as lacking of semantic support, leading to single point of failure and so on. To solve the problem that lacks of semantic support, many researchers introduce semantic Web technologies into Web service and add semantic information for the UDDI to achieve the Semantic Web service discovery.In order to solve the problems of the centralized Web service discovery, including a single point of failure, information overload, service failures and so on, this thesis designs a Chord-based semantic Web service discovery framework which consists of three layers that are regional community, communities of service and communities of interesting service and deploys the descriptions of service to Chord by using clustering method and hash function.This thesis discusses the related technologies and problems of semantic web service discovery, and uses OWL to describe shared ontology and domain ontology of services, as well uses OWL-S describes the information of services. OWL-S Profile is expanded by the descriptions on quality of service to improve recall rate and precision of service discovery.Combining OWLS-MX matching algorithm, this thesis proposes three-phase matching mechanism and query algorithm of services based on function and non-function of service. Functional match of services is mainly based on input and output parameters of the description of service. The matching process is that services are first matched based on the keywords, after the number of candidate services is reduced, semantic matching is executed, including the contextual match, match of custom rules of reference. User may choose the non-functional match of services as the complement of function match of service in which match of the quality of services can execute match of keyword and semantic. Three-phase service query algorithms, including query algorithm of locating region community, query algorithm of location interested community of services and query algorithm in interested community of services,can ensure that the user requests services in an accepted time.Finally, this thesis researches and implements the ontology module and the underlying Chord network in the framework of semantic Web service discovery and analyzes the performance of framework from the angle of recall rate, precision, query time. The results show that the mechanism of semantic Web service discovery has a higher performance of service discovery.