A novel prediction method of relevancy for focused crawling in topic specific search

A focused crawler is a web crawler which returns relevant web pages on a given topic in traversing the web. In order to determine topical relevancy, focused crawler use different classification schemes. We propose a novel method of predicting the relevancy of links in topic specific search. In this paper we focus on classification of links using decision tree induction and neural network classifiers to improve the performance of focused crawler. Our experimental results show that proposed approach has better performance than other related approaches.