Internet and database users expects desired output relevant to the input submitted, from the system, most probably may not get. It may be partial, maybe form, similar, sometimes exact. Lot of research solutions exists for this issue such as improved query processing techniques, automating and rewriting query as well as query reformulation, query expansion, knowledge base related search engine etc. Even then it lacks for user convenience or satisfactory level. In this paper as research a solution was analyzed and provided towards machine learning approach by taking the user feedback relevance from obtained output in an interactive way and processed to reach their convenience level.

Feedback relevance base semantic query processing efficiency analysis using neural networks

In our daily life World Wide Web has become an indispensable media. Its widely used to find a solution for a day to day basic activity issues such business, education, communiation, knowledge awareness etc but till their exists a question whether the web reached a full potential. Can it be till smarter and more intelligent? Answering these questions in concern Tim Berners-Lee the web inventor coined the term semantic web as extension of existing web and It is described as a giant global Web of data that is directly consumable and understandable to machines The Semantic Web is widely recognized as the most important enabling technologies that can help realize a smarter Web and enable Web intelligence. However the existing semantic caching schemas cannot be directly applied since the web sources typically have weaker query capabilities than the conventional databases. With this concern in this paper an analysis is made as an extension how the existing semantic information can be inferred intelligently in order to answer the user query further an experimental analysis is made over an academic system to check the optimality.

P. Mohankumar

Papers

Intelligence inference semantic caching and query processing