Research And Implementation Of Personalized News Recommendation System Based On Collaborative Filtering Algorithm

With the rapid development of the Internet technology, the users have entered the era of information overload gradually from the lack of information age. It is getting harder for users to find the interested information from numerous resources and the users' needs promote the development of personalized recommendation system. Personalized recommendation system has been more and more widely used in various fields, filtering a large number of content that users are not interested, predicting the user's potential preference.In this paper, the research content is the application of personalized recommendation system in the field of journalism. Although personalized recommendation for news in the theoretical research has made considerable development, there are still many problems to be solved, such as: data sparsity, cold start problem, scalability problem etc. Therefore, the focus of this paper is to solve the problem of data sparsity and cold start of personalized news recommendation system.(1)This article proposes a new collaborative filtering algorithm based on user preferences and project properties(IAUPCF). This algorithm is based on the traditional user item rating matrix, and it is integrated into the user's preference and project attributes. Finally the user item rating matrix is converted into the user item attribute rating matrix based on user preference(IAUPEM). Then the system calculates the nearest neighbor set of target user according to the IAUPEM, which overcomes the shortcomings of traditional similarity computing only relying on the user's score value.(2)It is suggested that an effective measurement method for predictor decision in this article. It can reduce the impact of "time effect" on the real evaluation of users by using the method, getting more accurate user rating trend, making the values after forecasting more closer to the real situation, and then improving the accuracy of recommendation.(3) The article has designed and implemented a simple personalized recommendation system for news based on collaborative filtering. This system uses the IAUPCF algorithm, which is able to accurately and timely give users news that they are interested in potentially.The IAUPCF algorithm can not only effectively make up for the shortage of the traditional collaborative filtering algorithm, improving accuracy significantly in the recommendation, but also provide a new research method for recommendation system for news.