MetaXplorer: an intelligent and adaptable metasearch engine using a novel ordered weighted averaging operator.

Search engines facilitate the access of information available on the World Wide Web. However, as the Web continues to expand, the portion of Web covered by each search engine is decreasing constantly. Metasearch engines address this issue by combining the results of multiple individual search engines and thereby, increasing the search effectiveness. This paper proposes a new model for metasearch, MetaXplorer, which is both intelligent and adaptable. This paper also proposes a novel Ordered Weighted Averaging (OWA) operator named Intelligent OWA operator, which is capable of handling the dynamic nature of decision making environment. The proposed Intelligent OWA operator is used for result aggregation in MetaXplorer, along with Fuzzy Analytical Hierarchy Process (FAHP). Furthermore, MetaXplorer analyses the documents returned by individual search engines instead of considering their ranks in search engine result lists alone in the aggregation process, and thus is intelligent. Subjective evaluation of MetaXplorer is provided by comparing it with previously proposed models. Also, the performance evaluation of MetaXplorer in terms of precision has been presented. The precision values of MetaXplorer are compared with three existing metasearch engines on the Web namely, Webcrawler, Excite and Dogpile. The results indicate that MetaXplorer performs better than the existing metasearch engines with the highest average precision of 0.6641, followed by Dogpile (0.5887), Excite (0.5723) and WebCrawler (0.5694), respectively. [ABSTRACT FROM AUTHOR]