Music recommendation algorithm based on user portrait

Based on the background of the big data era, information data is growing vigorously. When computers perform frequent calculations on massive amounts of data, user portraits can quantify data information through related algorithms and models, helping computers understand user needs and programmatically process various information. When computers can understand users, various related commercial applications will be greatly improved, both in terms of efficiency and accuracy. For different platforms and projects, the user role target states that need to be established are different. The more complete and accurate the model of the user portrait is, the more accurate the results of the recommendation system will be. This paper improves the music recommendation algorithm on the basis of predecessors, enriches the application of user portraits in the field of music recommendation, and solves some of the key problems existing in the current music recommendation algorithm, which has certain theoretical significance. The main task of this paper is to establish a portrait model of music users, and analyze the similarity of a large amount of data in combination with a clustering algorithm to obtain several user groups. Then, for each user group, a different recommendation list is obtained.