Generally, search engine retrieves the information using Page Rank, Distance vector algorithm, crawling, etc. on the basis of the user’s query. But it may happen that the links retrieved by search engine are may or may not be exactly related to the user’s query and user has to check all the links to know whether the needed information is present in the document or not, it becomes a tedious and time consuming job for the user. Our focus is to cluster different documents based on subjective similarities and dissimilarities. Our proposed tool ‘Web Search Miner’ which is based on the concept of user opinions mining, which uses k-means search algorithm and distance measure based on Term frequency & web document frequency for mining the search results. It takes an opinion from the user on the results given by search engine in different web documents. in response to the user’s multiple queries by downloading the pages in background, which saves the user’s time of searching a particular information and it gives the best results for the precise search by giving a mined search links. With the massive amount of information that is available on the World Wide Web, content mining provides the results lists to search engines in order of highest relevance to the keywords in the query.