基于大数据的个性化音乐推荐系统仿真

Simulation of Personalized Music Recommendation System Based on Big Data

笔者以million song dataset为音乐数据集,以taste profile为用户数据集,采用改进的基于用户喜好特征的协同过滤算法,实现了一个基于大数据的个性化音乐推荐系统。系统首先根据用户近期的听歌记录为用户生成听歌偏好模型,进而计算用户与歌曲的相似度,然后为用户推荐与其偏好最相近的歌曲。

This paper takes millisong dataset as music data set, taste profile as user data set, and adopts improved collaborative filtering algorithm based on user preference characteristics to implement a personalized music recommendation system based on large data. Firstly, the system generates a preference model for users according to their recent listening records. Then calculate the similarity between the user and the song, and recommend the song with the closest preference for the user.