Query sensitive comparative summarization using frequent terms set and semantic relevance based segments

Abstract

The exponential growth of World Wide Web has made it the prime database in the digital era for all information needs The enormous growth diverse dynamic semi structured and unstructured nature of web makes internet exceptionally difficult in searching and retrieving relevant information Search engines are helpful to some extent in locating the web resources For each query given by user search engines return thousands of URLs as search result which normally includes many irrelevant less relevant and redundant results This requires the user to further browse many web newlinedocuments to retrieve just a few lines of content of significance to the user Today there is a thrust for quick and immediate identification and retrieval of required information as expected by the users Automatic summarizers help the users to get the gist of the web page in few seconds without going through the entire content by applying web content mining techniques The query based summarizers process the entire document content newlineto identify and extract query relevant pieces of information at run time This challenge to the processing capacity of summarizers can be reduced by limiting the size of the textual unit that need to be processed newline newline