A Novel Approach on Focused Crawling With Anchor Text

Abstract

A novel approach with focused crawling for various anchor texts is discussed in this paper. Most of the search engines search the web with the anchor text to retrieve the relevant pages and answer the queries given by the users. The crawler usually searches the web pages and filters the unnecessary pages which can be done through focused crawling. A focused crawler generates its boundary to crawl the relevant pages based on the link and ignores the irrelevant pages on the web. In this paper, an effective focused crawling method is implemented to improve the quality of the search. Here, three learning phases are considered namely, content-based, link-based and sibling-based learning are undergone to improve the navigation of the search. In this approach, the crawler crawls through the relevant pages efficiently and more relevant pages are retrieved in an effective way. It is proved experimentally that more number of relevant pages are retrieved for different anchor texts with three learning phases using focused crawling.