A flexible and light weight hybrid Approach for discovery of owls Semantic web services

Abstract

ABSTRACT newlineService Oriented Computing (SOC) is one of the important developments newlinewhich took place after distributed computing era opened up. Web service is a newlinepremier technological development fueled by service oriented paradigm. The newlinetraditional discovery process using UDDI (Universal Description, Discovery and newlineIntegration) and WSDL (Web Service Description Language) could be improved by newlineusing semantic techniques. However, there is no light-weight and user-friendly newlinesolution which can readily be applied. This thesis focuses on three main problems newlineand their solutions. newlineand#61623; How to address the scalability issue in web service discovery? newlineand#61623; How to develop a matchmaking framework with a better precision rate? newlineand#61623; How to make the matchmaker more flexible and user-friendly? newlineThis thesis is a research effort, addressing the above mentioned three problems. newlineAs solutions, in this thesis three frameworks based on OWL-S (Web Ontology newlineLanguage Services) have been proposed. The S-FILTER framework aims at newlinetransforming the heavy-weight matchmaking process into a light-weight newlinematchmaking process, thereby addressing the scalability issue. The OWLS-FPH newlinematchmaker aims at developing a matchmaking system which could improve newlineprecision when compared to many existing matchmaking approaches. The newlineOWLS-FPTH matchmaker aims at making the system more flexible and newlineuser-friendly. The proposed OWLS-FLH matchmaker is an aggregation of the above newlinethree frameworks. newlineThe OWLS-FPH matchmaker considers Input, Output, Precondition and newlineEffect (functional parameters) of the semantic web service. The OWLS-FPTH newlinematchmaker adds flexibility to the matchmaking process. The OWLS-FPH newlinematchmaker works better only if the user is able to precisely define his/her request. newlineBut most of the time this may not be the case. Hence, to find a solution to this issue, newlinethe proposed framework is designed in such a way that it allows users to specify their newlinerequest using natural language. Then, to identify relevant services the user request newlineand the service text description are made to und