Repairing Broken Links Using Naive Bayes Classifier

Abstract

The Internet is an extremely useful resource for education and research. The Internet has been experiencing broken connections issue in spite of its concurrent services. Broken links are common issues stirring in the area of the web. Sometimes the page which was pointing from another page has been disappeared forever or moved to some other location. There can be many reasons for broken links such as the target website is for all time not available, the target website page has been detaching, the target web page was changed or altered and also has misspellings in the link. The broken link itself contains a lot of information such as URL, mark content, encompassing content close to naming content and the content in the page. Every one of these assets of information is valuable for recovering the candidate pages relevance for broken links. The system returns the ranked lists of highly relevant candidate pages on submitting a query which has been extracted from different sources. In this paper, we explore the expression that is used for the proximity (position) connection in the terms of the label and full text in order to extract relative (good and bad) terms through Naïve Bayes classification model. This solves the problem by providing non-identical terms to inquire multiple broken connections and also enrich the accomplishment as the terms that are closely identical show relevancy.