Research On Personalized Technology In Meta Search Engine

With the rapid development of the Internet, the increase of WEB information, and the demands from users to retrieve continuslly increasing. It becomes a difficult problem that users find the information they needeed in the vast informational ocean. At present,customers mainly using search engine to retrieve information. However, the shortage of search engine is which provides low quality service. The reason is that user has their own interests and requirements, and search engines do not consider their own factors, search engine can not me as a center, it just uses the index method based on general intent.Therefore, the article bases on complex analyse about search engine teachnology at home and abroad,it deeply studies that how to improve users personalised service.First of all, it introduces some realative basic knowledge of independent search engines and meta-search engine, it finds the shortage of the present search engine model through search lots of references. It will bring users habit characters into the new model. We create a personalized Meta-search engine base on users habits and introduces the process of the model detailedly, This model in order to better achieve the user's personalized service, and improve the user'recall and the precision.Secondly, after analysing the advantages and disadvantages of Web users clustering algorithm which the Boreges and Chen, Levene and others have proposed, it proposed a new Web user clustering algorithm which based on the user interest features. Through users interest and browse pathes to delete the information, reduce the complexity. Use the fuzzy set theory to chuster users, so The operation of the algorithm is to improve the clustering speed, as much as possible to meet the requirements of different users of the search.Finally, to improve the traditionl recommendation technology, proposed a technology based on a Web user clustering recommendation, this technology in order to reduce the time and improve browsing efficiency, enhance the overall performance of recommended technology, it is a hot topic of current research. that Whatever theory analysis and experimental results show that the recommended technique than existing traditional recommend technology performance more superior.From the view of personalized characteristics, the purpose is to improve the performance of overall system, improve search accuracy, save the user's retrive time.