In order to improve the accuracy of news recommendation, this paper established a comprehensive user interest model. First,this paper established a stable user interest model based on user browsing habits .Then,the paper also advanced freshness-based tentative recommendations on the basis of news timeliness and mainstream to get the user's temporary interest model. Finally,the paper combined these two models to establish a comprehensive user interest model. Experimental results proved that the proposed method can recommend specific news articles that best meet the user's reading preferences from a large number of the latest published news.
A Microblog User Interest Model Based on Participated and Not Participated Data
