METAFUSION: AN EFFICIENT METASEARCH ENGINE USING GENETIC ALGORITHM

Abstract

World Wide Web is a dynamic source of information which is expanding its content at a staggering rate. Individual search engines are not able to handle the exponential nature of web. Hence meta-search engines are used to solve the problem of low web space information coverage rate of individual search engines. A meta-search engine is a kind of search tool that dynamically dispatches user query to the underlying search engines, hence providing parallel access to multiple search engines and then aggregate the results to present single consolidated result list to user. In this research work , a novel meta-search engine, MetaFusion , has been proposed. The proposed algorithm uses Fuzzy AHP along with Genetic algorithm to get more comprehensive and optimized results. Fuzzy Analytical Hierarchy Process reflects human thinking and addresses the uncertainty of information while making decisions in MCDM problems. Genetic Algorithm(GA) is highly robust and self- adaptive algorithm , hence, solves the more complex problems in optimum manner.GA uses average weight of document in underlying search engine as fitness function for merging results. Experimental results shows that relevancy of returned results by MetaFusion is more than several existing research Metasearch engines. The precision of proposed model MetaFusion comes out to be more when compared with available research metasearch engines i.e. Dogpile ,Infospace and PolyMeta.