Research On Semantic WEB Services Discovery And Its Application

The creation and development of Web Services make people can use application in module any platform, anywhere, anytime. However, users don't know web service is exist before use it, so we need find them, that's means Web Services Discovery. At present, UDDI solves problems: finding and announcing web services in web architecture. In long views, using WSDL to describe web services has less semantic information and make it's hard to adapt to web services' increase and automationThe target of web services discovery is making discovering process efficient and automation. The creation of semantic web solves these two questions. Integrating semantic web and web services together can make discovery technology efficient and automation. We use a language which based on XML and include semantic information to describe web services in semantic web services. So web services understanding each other contents, functions and properties become possible. And offer a technical support for discovery, execution, combination and monitor.This paper investigates discovery technology of semantic web services, starts with semantic web services discovering process, design the model of semantic web services discovery. In this model, there are two greatly sub-module, they are extended UDDI and ontology database. In this model, we match services one layer after another, there are three layers together, they are: submitting service request, constructing services set which similar to request profile and make sure the matching services . set. In the process of login and requesting service, we use OWL-S as the web services' semantic description and communication language.This paper constructs candidate services set and make sure match services set through estimating semantic web services' competence. In the process of constructing candidate services set, we filter web services layer after layer. Firstly, we match web services' category bag. In this step, we can filtrate lots of web services which not suit to the service query. Secondly, we match inputs, outputs, precondition and return of querying service and announcing services. Lastly, we match quality of services. So this method is efficient in discovery of services.This paper designing and realizing the system, so validate the model is usable and validity.