Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Computer network > General issues > The application of computer network

Finding Web Services Based on Clustering Probabilistic Semantic Approach

Author HanRui
Tutor ZhangWeiFeng
School Nanjing University of Posts and Telecommunications
Course Computer Software and Theory
Keywords Web service Web services Matching Probabilistic Latent Semantic Analysis semantic clustering similarity
CLC TP393.09
Type Master's thesis
Year 2012
Downloads 19
Quotes 0
Download Dissertation

Recently,Web services have been increasingly used to integrate and build business applications on the Internet.Once a Web service is published and deployed,climents and other applications can discover and invoke it.With the incredibly increasing number of Web services on the Internet,it is a challenging issue for service users to discover desired services that match their requirements in service-oriented computing.Currently,UDDI is a standard for publishing and discovering of Web services,and also provide keyword searches for Web services.However,the search functionality is very simple and fail to account for relationships between Web services.In order to efficiently finding Web services, the semantics in Web serivces should be considered.Based on the current dominating mechanisms of the discovering and describing Web services with UDDI and WSDL, the proposed method utilizes Probabilistic Latent Semantic Analysis to capture semantic concepts hidden behind words in a query and the advertisements in services so that services matching is expected to be carried out at concept level.The mainly work and contribution come out of the thesis are:1 This study compares the LSA, PLSA, LDA models and clusters services semanticlly then the semantic similarity of a query and Web services is measured within the related semantic cluster2 Computing the similarity of service names, description from Web pages and operation names which have different weight in similarity compution and finally get the semantic similarity between Web services.3 Utilizing the LSA analysis to initialize a PLSA model and comparing the PLSA model with the LSA-based PLSA model

Related Dissertations
More Dissertations