Design And Implementation Of The Personalized Chinese Search Engine Based On Secondary Sort

With the rapid development of the network and computer system performance, Internet has become the global largest information repository. Users are faced with the problems of "information overload". How to provide efficient information retrieval service in such a huge information resources and to help users quickly find needed information in the ocean of data are urgent need to solve in search engine. One of the ways to solve the above problems is to develop the personalized search engine. To realize the personalized service has become a hot topic in the study of the search engine technology.This thesis mainly introduces the research status of personalized search engine and the shortage of existing search engines, combining with the first and second generation of search engine technology and analyzing the Web pages when users using a search engine. A kind of personalized search engine model is put forward, which is called the Personalized Chinese Search Engine Based on Secondary Sort with the real-time analysis by the client, instant response feedback and instant reordering. A search engine named Myso is implemented based on the model. The main works are as follows:(1) On the basis of studying the working principle of personalized search engine, the Myso search engine that based on secondary sorting personalized search engine system model is designed and implemented.(2) By learning the term frequency/position weighing algorithm, it is to design the formulas and algorithms of calculating Web Rank which are based on the word "e Weight" of Chinese word segmentation system "ICTCIAS", calculate initial sequence of web pages and simplify Myso search engines.(3) In order to obtain the interest of users, this thesis adopts the technology which implicit collects users interests, to propose a method of instant updating the users’ interest by tracking users’ behavior, and on this basis, combined with the web page classification technology, this thesis proposes a personalized sorting method named the personalized ranking method based on secondary sorting.