A NOVEL APPROACH FOR CONTEXT BASED

The WWW (World Wide Web) is a huge repository of hyperlinked documents known as web documents, accessible through Internet [1]. It is an information sharing model that is built on the top of the Internet. WWW is a continuously growing collection of hypertext documents with its estimated indexable size of at least 50 billion web documents stored on thousands of web servers world wide [2, 3]. Owing to exponential growth of information, World Wide Web has become an important source for searching the required information. It operates on the Internet’s client-server architecture. But, these web documents are not organized as books in library and no central catalogue for the same is available. Even knowing where to look for information using uniform resource locator (URL) is not a guarantee that it will be retrieved. Therefore, it is very difficult to search the desired information from such a huge collection of web documents in an efficient manner. Searching the huge WWW repository has become a challenging problem. As a solution, several information retrieval (IR) tools have been developed. These tools are divided into following three categories: Web directorie