Research And Implementation Of Personalization Of Agent-based Intelligent Meta Search Engine

With the inflation of information resources and the enhancement of users’ demands,traditional search engines gradually showed the problems of low coverage and servicesnot intelligent enough. Thus the intelligent meta-search engine which could effectivelyimprove the coverage and accuracy was put forward and got high attention. This papermainly researches the personalization function of Agent based intelligent meta searchengine and takes advantages of Agent technology to realize the active learning of searchinterests and personalized filtering. It will support users to get more relevant and highquality search results.This paper mainly introduces the research and implementation of personalizationfunction of Agent based intelligent meta search engine. Firstly this paper shows thedeficiencies of current search engines. Then it gives an overview of the relevant theoriesand technologies. Next this paper puts forward the Agent based intelligent meta searchengine which could solve the deficiencies, including the framework of the Agent basedintelligent meta search engine, the retrieval results synthesis model, the Agent structuremodel and Agent life cycle. Next, this paper studies the design of personalizationfunction, provides the process of interests mining and queries analysis. Then it designsthe recommendation mechanism of search results based on public interest blackboard,the recognition mechanism of complex inquires based on dynamic learning process andthe databases. Next, this paper provides the concrete realization of the personalizationfunction, including the realization of recommendation mechanism of search resultsbased on public interest blackboard and the recognition method of complex inquiresbased on dynamic learning process.According to the experiments to the relevance of search results and the effect ofrecognition of complicated queries, the personalization function of Agent basedintelligent meta search engine put forward in this paper could studies the users’ searchinterests effectively, improves the relevance of search results obviously, identifiescomplex queries accurately. It could give full support to improve the efficiency of users’information retrieval.