New media popular music recommendation system based on machine learning algorithm

With the development of new media and AI, music recommendation systems have become essential in meeting diverse user preferences. Traditional collaborative filtering methods neglect music content, leading to less accurate recommendations. This paper proposes a machine learning-based popular music recommendation system that addresses these limitations. By analysing user preferences and applying the maximum entropy model (MEM), the system integrates and processes music data more effectively. Experimental results show that the new approach enhances recommendation accuracy by 7.5%, making the system more intelligent and user-friendly. This study contributes to improving music recommendation performance in the era of new media.