Research On The Fuzzy Clustering Algorithms In A Personalized Research Engine

The invention of network brings a great change to people. Information gives people not only happiness, but also some problems about how to use it. Now a days, Search engine is the primary tool for people to getting information on Internet. The search engine based on keywords can't fit to the complex Web. Obviously, the present problem on search engine is how to return more intelligent searching result for users.This paper introduces a new Personalized Search Engine Model, which is base on the GVFCMdd algorithm, in order to adapt to the need of Web searching, to ameliorate the searching method base on keywords totally, and to improve the preciseness and recall of search engine.This system analyses Web documents with fuzzy clustering method. In addition, this system constructs a personalized model for every user to describe the user's interesting. The model can make search engine return more personalized searching result for users.With current search engine technology, adding fuzzy mathematics conception, using the clustering method and generation algorithm idea in Data Mining technology, to overcome the shortcoming of FCMdd algorithm, which is dependence on initialized spots, this paper presents a new fuzzy clustering method based on generation algorithm, GVFCMdd in short.This system has realized initial intelligence of search engine, and improves the preciseness and recall of search engine to a degree.