The Research Of User Interest Point Mining Based On Knowledge-Base And Text Classification Algorithm

In recent years.with the rapid development of the internet.people can get the information they need through the internet.Currently search engine has become an important search tool,but the information is the same for different users.not processing for the individuals,so it ignores the user’s interests,which can not meet the user’s real individual needs.The mining of the user interest points is to dig out the user’s interest point from the user’s browser history,its results directly reflect the accuracy and effectiveness of personalized service.this paper carryed out researches on the user interest mining.This paper analyzed the related algorithms on the user interest point mining.and according to the limitations of existing algorithms.an algorithm basing on knowledge-base and text classification algorithm to mine user interest point is proposed; this paper is in English corpus environment.firstly,this paper created Wikipedia-based knowledge by using Lucene.and then classified the user input keywords and-user input URL.finally projected the users interests.in the keywords classification.this paper proposed a combined classification algorithm based on the co-occurrence of words and WordNet expansion:in the URL classification.this paper proposed a context extraction method based on the block of the page.and also proposed a DFSD characteristics extraction method;in the projection of user interest.this paper proposed a method based on the context.the user candidate interest points are mapped to a interest point, which reflected the real interest point.finally the experimental results reflected the effectiveness and accuracy of the algorithm.