Service Description Model Of Service-oriented Computing Research

The service description is a set of documents describing the service interfaces andsemantics; it uses service semantics, service relationships to represent a service, anddescribes the services, service interfaces and exchangeable informations in a format thatcan be processed by machines. Service description is the premise of service discoveryand service matching. The service discovery helps service consumers to find resourcesand services of solving specific problems, and helps service providers to enable serviceconsumers to note the services they provide; service matching helps to find thesimilarities between the service entities in order to support synergic services.In recent years, the practices of service discovery, service composition andautonomic computing in Service Oriented Computing show that a semantic conventionabout services is particularly important, semantic description of services providesmachine intelligibility for the potential complexities of Web Services.The representation of service description model goes through from the text toknowledge; OWL-S, which is based on OWL (Ontology Web Language), becomes tobe the standard ontology to represent service description model; but the OWL-S servicedescription model only contains static semantic, doesn’t consider the evulation ofservice itself and the domain it belong to in service lifecycle. So it can’t reflect thechanges classes and their relationships of the domain that the service belongs to, and thechanges of QoS. To solve the problems, this paper made some innovative andexploratory researches on service description model based on OWL-S, the maincontents are:(1) Extend the OWL-S ontology with dynamic semantic, form a service modelwith rewriteable description logic. Specifically, add fuzzy classe model to extend theclasses and class relationships of the service ontology, and add QoS model to reflect thestate changes of the service.(2) A hybrid service matching algorithm has been designed for the extendedservice model, unlike traditional service matching algorithm, this algorithm is designedto match different content with different method to refine the matching granularity, and it takes full account of the fuzzy matching and QoS matching of service.(3) A semantic web service discovery system has been designed, which includesservice query and matching components based on semantic compared with traditionalsystems, it simplfies the complexity of the service registry center, at last I have verifiedthe features of the extended service model and matching algorithm with the servicediscovery system.