Effective Performance of Information Retrieval by using Domain Based Crawler

Abstract

World Wide Web continuously introduces new capabilities and attracts many people[1]. It consists of more than 60 billion pages online. Due to this explosion in size, the information retrieval system or Search Engines are being upgraded day by day and it can be used to access the information effectively and efficiently. In this paper, we have addressed Domain Based Information Retrieval (DBIR) System. In this system we crawl the information from the web and added all links to the data base which are related to a specific domain. It simply ignores which are not related to that domain. Because of that we can save the Storage Space (SS) and Searching Time (ST) and as a result it improves the performance of the system.It is an extension of Effective Performance of Web Crawler (EPOW) System [2], in which it has two Crawler modules. The first one is Basic Crawler. It consists of multiple downloaders to achieve parallelization policy . The second one is Master Crawler, which is used to filter the URLs send by the Basic Crawler based on the Domain and sends back to the Basic Crawler to extract the related links. All these related links are collectively stored into the database under a unique domain name.