Web service discovery is an important activity in services computing paradigm. Omnipresence of internet and proliferation of web services provides an opportunity to many service providers to create, locate and access the business services. Recommending pertinent web service to the client’s requirement is a complex task as numerous functionally similar web services are readily available to meet the demand. Furthermore, the conventional method does not support the client’s non functional requirements. To alleviate the difficulties, web service search engine has been introduced to assist the user to search the appropriate web services. The proposed approach consists of functional and nonfunctional evaluation for computing the similarity measurement, and normalizing the QoS values respectively. The WSDL parser extracts the Meta data contents from the WSDL file and it will be preprocessed. Soon after preprocessing, appropriate terms will be identified. To find the similarity between two web services, the semantic distance between terms will be calculated using Normalized Google Distance (NGD). The degree of similarity between web services can be calculated by using bipartite graph algorithm. The score value will be computed based on aggregation of functional and non functional measurement values. For a user request, a list of candidate services will be provided to the client based on degree of matching with the search query from higher to lower order. The proposed system has been tested with real world web services and the test result shows better precision and recall values, which outperforms existing approaches.