S-mse: Asemantic Meta Search Engine Using Semantic Similarity and Reputation Measure

In order to increase web search effectiveness, Meta search engines are invented to combine results of multiple search engines as a result of larger coverage of indexed web. Meta search engine is a kind of system which is useful for internet users to take advantage of multiple search engines in searching information. Recently several approaches were developed using ontology and ranking measures. Accordingly, Meta search engine is developed here using ontology and semantic similarity measure. In order to bring semantic in keyword matching, a semantic similarity measure (SSM) is developed. Here, every concept sets are matched with the title sets using SSM that consider the hyponyms and hyponyms of the keywords presented in the title sets. Along with three different ranking measures relevant to contents, title sets and raking value given by the standard search engines are effectively combined to improve the effectiveness. Finally, the experimentation is carried out using different set of queries and the performance of the meta-search engine is evaluated using TREC-style average precision (TSAP) measure. The proposed semantic meta-search engine provides 80% TSAP which is high compared with existing search engine and meta-search engine.