Abstract
Discovery of semantic Web services is a heavyweight task when the number of Web services or the complexity of ontologies increases. In this paper, we present a new logical discovery framework based on semantic description of the capability of Web services and user goals using F-logic. Our framework tackles the scalability problem and improves discovery performance by adding two prefiltering stages to the discovery engine. The first stage is based on ontology comparison of user request and Web service categories. In the second stage, yet more Web services are eliminated based upon a decomposition and analysis of concept and instance attributes used in Web service capabilities and the requested capabilities of the client, resulting in a much smaller pool of Web services that need to be matched against the client request. Our prefiltering approach is evaluated using a new Web service repository, called WSMO-FL test collection. The recall rate of the filtering process is 100% by design, since no relevant Web services are ever eliminated by the two prefiltering stages, and experimental results show that the precision rate is more than 53%.
Highlights
Semantic Web has been a popular topic of research since its introduction by Berners-Lee et al in 2001 [1]
Our semantic Web service discovery framework focuses on Web services, goals, ontologies, and mediators that are semantically described based on the Web Service Modelling Ontology (WSMO) [11] model and using the F-Logic [12] language as implemented in the FLORA-2 [13] logic system
OWLS-TC, which mainly considers input and output parameters, is applicable for approaches that deal with OWL-S Web services descriptions, and approaches which employ Semantic Annotation for WSDL and XML schema (SAWSDL) Web service descriptions use the SAWSDL-TC test collection
Summary
Semantic Web has been a popular topic of research since its introduction by Berners-Lee et al in 2001 [1] Based on this idea, automation of many tasks on the Internet is facilitated through the addition of machine understandable semantic information to Web resources. This paper presents a new logical framework and two prefiltering strategies to improve the speed and accuracy of automated Web service discovery. Our discovery framework is based on the WSMO conceptual model for semantically describing user requests (goals), Web services, and domain ontologies. Goal capability descriptions such as inputs, outputs, preconditions, and postconditions (effects) are compared with advertised Web service capability descriptions in order to determine whether they match or not. Our second prefiltering stage uses a new technique of extracting attributes and concepts of objects utilized in the goal and the Web. Scientific Programming service pre- and postconditions.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
