Meta Search Engine using Multi-Objective Partial Rank Aggregation: Application in Ranking WebPages

Although there are hundreds of search engines no single search engine can satisfy all web users and can be considered broadly acceptable that Sufficiently comprehensive in its coverage of the web moreover they consist the “spam pages” when a web page gets an undeservedly high rank. Therefore, a robust technique for Meta Search Engine is required that can effectively combat “spam pages”, a serious problem in Web searches. When some top ranked WebPages are extracted from different Search Engines, they differ each other, and the problem become more difficult to aggregate them. In this article a multi objective genetic algorithm based rank aggregation method has been proposed where some partial rankings are integrated in an unbiased way. The total distance from the reference ranking to the input rankings is minimized as the first objective. For distance calculation the scaled footrule distance is used. The standard deviation between those distances is minimized as the second objective in order to avoid biasness toward a particular input ranking. The proposed method has been applied on some most-usable Search Engines. Thereafter, the resultant ranking is compared with the other rank aggregation methods to check how much it is applicable to reduce the spam pages.