Research On Adaptive Teaching And Research Information Recommendation Model

Along with the rapid development of the network information resources, more and more people are facing difficulty in choosing and digesting information. The distribution universality and disorder of Internet information resources leads to search information more difficult. Because the search engine has universal nature, it is difficult to satisfy the different queries of the different background, different purpose and different periods of the user. The information needs of users are usually with personalized characteristic, which are relatively stable and also change with the time. So users look forward to information service software that can meet personalized information requirement.Personalized recommendation information retrieval technology has become one of the hot research fields. So-called personalized service is that choosing different services models and providing different service contents to different users, the key of personalized service is to known interests of users, and then accurately build interest models of users. How the interest model reflects the user interest decides the quality of personalized recommendation service provided by software system.This paper puts forward a method based on results of cluster analysis model of personalized recommendation, first, introduces the major web mining technology and analyzes the user modeling technology, and puts forward a user interest mining process model on complementary analysis the information content of teaching and scientific research, supplementary analysis the information of users browse behavior. Then, research on the technology of the page document said, including: text vector space model, the choice and will feature text page for vector space model of structured format. Page and the user can use the model is used to the similarity functions in similarity calculation. And reuse of information fusion algorithm to reorder these search results, cluster analysis results of different types of documents, adjusts weight after recommendation. Finally, do a more comprehensive experiment for the improved method. According to the teaching and scientific research information push the analysis of experimental results show that the proposed model can be accurately describe the user's interests; the personalized recommendation has practical value. Based on the results of cluster analysis personalized recommendation model can improve recommend effect. This is the point of the innovation of the user behavior records into user feedback information in the model for the user, adaptive recommended for teaching and scientific research information.