On Selective Result Merging in a Metasearch Enviroment

When a query is passed to multiple search engines, each search engine returns a ranked list of documents. The problem of metasearch is to fuse these ranked lists such that optimal performance is achieved because of the combination. There are two parts of the metasearch process. The first part is to select which search engine results are to be merged. The second part is the actual process of merging. In this paper we propose (1) a strategy for selective merging of results from a metasearch engine and (2) a heuristic for handling missing documents in result sets to be merged to improve the result-merging process Our experimental results will show that our proposed strategy for selection before merging coupled by incorporating it into the BORDA method improves the performance of merging.