Examining the Feasibility of Metasearch Based on Results of Human Judgements on Thai Queries

This paper provides the results of public Web search engine evaluation based on queries written in Thai. Statistical testing shows that there are some significant differences among engines. Besides comparing the effectiveness of Web search engines, the returned results are compared in order to illustrate the relation and overlap among these results. The results reveal that the majority of returned results are quite unique. Since the results among engines differ greatly, this encourages the use of metasearch approaches to combine best search results from different engines. We examine metasearch models based on the Borda count voting scheme. We also propose the use of evolutionary programming (EP) to optimize the weight vector used by the Borda count algorithm. The results show that the use of metasearch approaches produces superior performance compared to any single search engine on Thai queries.