Web service discovery is an important process in services computing paradigm. Due to the enormous number of service providers available for a given functionality, identifying the suitable service to the client requirement is a challenge. The traditional discovery mechanism supports syntax-based discovery, and it limited support on semantic discovery. Web service registries and portals does not support client?s QoS requirements. The general purpose search engine such as Google, Baidu may provide the required results for a search query. Since these search engines are mainly designed for retrieving web documents, they are suitable for discovering web services. Traditional methods based on Term frequency (TF) and Inverse Document Frequency (IDF) will not suitable for web services. The repetition of terms and finding the richness of the document will not be applicable for finding appropriate web services. Due to the fact that WSDL contains much less information than web pages, these service discovery mechanisms are not applicable for discovering web services. To alleviate these difficulties, the Quality Driven Web Service discovery Framework QWS-Search has been introduced. The system consists of functional and non functional computations so that the returned results is semantically indexed and list them based on quality. The system QWS-Search has been tested with more than 2000 real world web services, and the result reports that the proposed system which outperforms the existing implementations in terms of precision and recall values.
展开▼
