Optimal Query Generation for Hidden Web Extraction Through Response Analysis

Abstract

A huge number of Hidden Web databases exists over the WWW forming a massive source of high quality information. Retrieval of this information for enriching the repository of the search engine is the prime target of a Hidden web crawler. Besides this, the crawler should perform this task at an affordable cost and resource utilization. This paper proposes a Random ranking mechanism whereby the queries to be raised by the hidden web crawler have been ranked. By ranking the queries according to the proposed mechanism, the Hidden Web crawler is able to make an optimal choice among the candidate queries and efficiently retrieve the Hidden web databases. The Hidden Web crawler proposed here also possesses an extensible and scalable framework to improve the efficiency of crawling. The proposed approach has also been compared with other methods of Hidden Web crawling existing in the literature.