Strategies for enhancement of natural language based web service discovery mechanisms

Abstract

Web Services are self-contained, modular, distributed, dynamic applications that can be described, published, located and invoked over the Internet. Web services are said to be an application services, which are made available from a business web server targeting the web users or other web based programs. The web service context is easily understandable in large applications that need appropriate web service for quick data fetching and retrieval. Natural Language Processing (NLP) is a category of Machine Learning (ML). However the existing NLP approaches lacks semantic repository that can define a direct meaning to the query keywords. The research proposes a framework that identifies appropriate web services with semantic understanding using query categorization. The query categorization is analysed using Machine Learning algorithms. The entire research has been carried out with a corpus of query keywords and is used for query understanding. The aim of the research is to make understand how the query can be built for accessing the appropriate web service against the corresponding domains. The proposed framework enumerated algorithm for Ontology Classification phase, Input specification Extraction, Output specification phase and subject specification extraction phase. A query optimizer and classifier are also being designed. The approach is significant as it converts a natural language query into exact query for retrieving information. The improvement is shown in terms of time, precision, recall, F1-Score and Support. The domains covered are Education, Travel and Economy. The architecture is significant as it covers all the phases right from query search to retrieval using web services. An improved mechanism for storage and access and retrieval is suggested by using hash based indexing approach. newline newline newline