Abstract
Document clustering is an important technology which helps users to organize the large amount of online information, especially after the rapid growth of the Web. This paper focuses on semantic document clustering method and its application in search engine. We proposed a multi-agent based information retrieval system to enhance the search process. The agents retrieve the results of Web search engine and organize the results by clustering them into different categories for a given query. We utilized WordNet ontology and several approaches to cluster results in appropriate category according to WordNet synsets. The experiment shows that semantic clustering work better than original clustering.