Indian Journal of Science and Technology

Abstract

Objective: This work discusses about web search based on content present in the web page. Web 3.0 in this field has been a major invention done till this date. Yet semantic web proposes to be a good choice for its enhancement towards searches based on intent of user. There has been lots of techniques proposed so far still methods are been searched based on this concept. Methods: Surfing the internet for relevant information as needed by the user has therefore become an issue for users. In order to overcome the problem of information deluge for relevant information retrieval, there is a need to devise a way to search data based on user’s intent. This is where semantic search comes into play. Findings: In order to improve quality of search correctness, semantic search tries to interpret the context of data provided along with user’s intent, location from where that has been looked for, variation of words etc. As a part of our work we have proposed a model using natural language processing that aims to classify data based on semantic relevance to user query. The well-known concept of NLP permits the machines to derive the purpose or the meaning meant by humans or natural language input. The aim is to provide exact needed result to the user in place of making him to look for it all over the URLs given as result. Improvements/Applications: Tools based on this semantic knowledge will increase relevance of search engines along with the less list of documents o go through by the user or surfer for a specific result. 
Keywords: Information Retrieval (IR), Natural Language Processing (NLP), Semantic Web, Semantic Relevance, Stop Words