Research And Design Of System Based On Personalized Search

In the university's information construction, web learning platform is the core part. It plays an important role in assisting user's study and enhancing their self-educated abilities. Site search in traditional learning system is based on title or keywords, sometimes users want to know the inner content of the resource, at that time, full-text research is to be needed. This paper is proposed in this situation. According to the architecture and realization of search engine, this paper will realize full-text research in a simple system and then optimize it by personalization service, so that this search engine can give different results to different users according to their interests, then users can find what they are interested in at the first time, making this system assist user's study better. The main research works in this thesis are as follows:First, this paper researched the format of how to store the information, finding that it is viable of storing it into html format. Then the resources can be regarded as web pages, we can make use of the reality of the search engine, and can make use of Lucene to develop it.Second, this paper researched search engine, personalized search engine and some free jar package for development, such as Lucene, HtmlParser and Chinese segmentation, while Lucene is used to develop search engine and HtmlParser is used to analyze Html. Based on the above, this paper designed the architecture of the personalized search engine in this system.Third, this paper researched the most important questions about personalized search engine:the build of user interest model and the improvement of page ranking. Both of them are designed by considering many factors and the actual situation.Fourth, this paper designed a system based on Struts, testing personalized search engine based on it.