News Recommendation Based on Content Fusion of User Behavior

To address the problem that content-based recommendation algorithm lacks diversified user interests, a news recommendation algorithm based on content fusion user behavior is proposed. Firstly, the content-based algorithm is used to improve the computation of user interest weights by considering the effect of time on user interest and to discover users' own preferences. Secondly, considering the time utility and the correlation between the popularity of the news and users' interests, finding users' nearest neighbors using content-based and improved behavior-based mixed similarity to discover their potential preferences.Finally, the user's own preferences are fused with potential preferences to generate a recommended list. The experimental result shows that compared to the traditional recommendation algorithm, the improved algorithm has a maximum increase of 9.80% in recommendation accuracy, a maximum increase of 10.12% in recall, and a maximum increase of 6.23% in diversity. The experimental result shows that the improved recommendation algorithm effectively improves the quality of recommendation, enrichs diversity, and has good recommendation effect.