Distributed Ontology-Driven Focused Crawling

Abstract

Focused crawlers are programs designed to download web pages which are relevant to specific topics. Using information gathered at running time, focused crawlers explore the web following promissory hyperlinks and fetching only pages which appear to be relevant. These crawlers are receiving increasing attention because they favor the construction of vertical search engines, allowing users to focus on specific topics of information, providing higher accuracy and reducing computational costs involved in query processing. In this article, we introduce an efficient focused crawling strategy which considers a number of distributed focused crawlers which recover relevant pages to a given knowledge domain. We propose an ontology-based knowledge representation approach to drive the crawler to specific segments of the web. Experimental results with actual samples of the Web show the feasibility and efficiency of our strategy.