Each day more and more documents and pages are created on the web. As each new piece of information competes for our attention, it becomes increasingly difficult to sift the pertinent pages from the rest of the information. “Information filtering recommenders” look at the syntactic and semantic content of items to determine those that are likely to be of interest or value to a user. A content-based search engine examines the content of the document and checks its correspondence with the search query. A pure, content based recommendation system is one in which recommendations are made for the user based solely on the content of items which that user has rated in the past. Online readers need tools that can help them cope with the mass of content available on the World-Wide Web. Recommender Systems are agents that people can use to quickly identify content that is most likely to interest them. The system that we have designed aims at integrating the information retrieval system with the collaborative-based filtering approach. Unlike most other search engines, this system looks even at related terms. For instance, it will be able to recommend technical papers to its users based on the similarity of the user query in the paper. It will also recommend papers that other users with similar search interests have earlier looked up.