Research On Ontology-Based Semantic Web Services Discovery And Selection

With the rapid development of Internet, network application has already gone deep into the very aspect of society life, Web services can be acquired sharply, at the same time, the on-line information increases exponential, so it become more and more difficult to seek, visit, represent and maintain services to the customer coming from each field. The problem, "abundant data and indigence knowledge", is becoming more and more outstanding. It is valuable to efficiently locate desired web services and select the best one from a group of functionality-similar web services, while these just are the task of web services discovery. However, the WSDL/UDDI of web services discovery technology is done through keywords matching based on the syntactic description of web services. Such web services discovery technology can't capture the semantic information of web services and is lack of intelligence, so it can't satisfy the desire of people. In order to get a better result of service discovery, a more efficient and perfect web services discovery technology is needed. The purpose of semantic web is to extend the current web, make the information on the web easier to be understood, or managed by machine. So the research of semantic-based web services discovery and relative problems is a hot topic. The researches show that the use of semantic description and ontology of Web services can improve the result of service discovery.Service matching is the key issue of service discovery. Service matching is done by compared the desired web services description of requester with the service description of provider published,and judging whether the latter satisfies to matching or not. When the service description published and requested is exactly alike, it means this is the most perfect service matching, however, this kind of strict definition make the recall and precision of service discovery all very low, even the failure for service matching will be caused. A good algorithm of matching should have the certain loose degree, according to a certain similarity degree, so matching can be realized according to the need of service requester and the description of the service provider, so as to attain the service satisfied need possibly.The selection of the web services,means when the service requester facing a large quantity likely services, it can be use to do services choice according to correspond criterions so as to raise the accuracy and the credibility of the service selection.The semantic model of the web service description, OWL-S is studied in this thesis, and...