Research And Application Of Personalized Information Service System Based On Hybrid Recommendation

With the popularity of the Internet and the development of global information, the Internet has become the main way of people getting information; at the same time, resources of information are becoming grown explosively. The users often difficultly find the valuable information, and some information which is rarely concerned about will be easily become isolated. The personalized information service system can solve these problems effectively, it analyzes the user's interest and browsing record, and then recommends them information that they are interested in and help them find the information they really need. But the current personalized information service system is not mature with some general drawbacks existing in recommending efficiency, quality and so on.This thesis studies the personalized recommendation technology, and focus on content-based filtering and collaborative filtering technology to study and compare their strengths and weakness and to explore the combination of two recommended techniques for mixed-recommended ideas.In the study on the personalized recommendation system, we combined the personal information, user's score and query keywords together to establish the users' interest model, and calculate the similarity between them, and then cluster users in the off-time. Thus the online recommend time is significantly reduced and the response time of the recommendation system is improved.Considering the quality drawbacks of current personalized recommendation system, we calculate the items'similarity to solve the problem of cold-start, then propose to combine with content filtering technology to forecast the missing values in user-item matrix in order to solve the data sparse problem and improve the quality of recommendation. Lastly, we design and implement a personalized recommendation system of video site based on the hybrid model.