Research On Music Recommendation System Based On Emotion Analysis

With the rapid development of Internet technology,the number of music in the network is increasing day by day,and the research of music recommendation system is also advancing.Recommending music content according to users’ emotional needs has gradually become one of the main research directions of music recommendation system.However,at present,the recommendation function of traditional music software has some problems,such as low recommendation accuracy and weak user pertinence.These problems may lead to the music recommendation results unable to meet users’ needs,and then affect the use experience of the recommendation system.In order to solve the above problems,from the perspective of emotion analysis and recommendation algorithm,researchers combine music emotion analysis technology with music recommendation algorithm,propose music recommendation algorithm based on emotion analysis,and apply the algorithm to music recommendation system.Music recommendation system is a common music resource filtering method,but for the personalized needs of users for music,the traditional music recommendation system has some defects.This paper first optimizes the music emotion analysis model,collects and analyzes the user emotion and music emotion,and then carries out music recommendation according to the analysis results.After the experiment proves that the algorithm is feasible,the music recommendation algorithm combined with emotion analysis is applied to the music recommendation prototype system,and then completes the implementation of the prototype system.The main work of this paper includes:(1)Research on music emotion analysis model.The application method of the model is improved on the traditional music emotion model,and the use method of Hevner emotion ring is optimized.By using two emotion types in the emotion ring to describe the same music,the classification method of music emotion is improved to make it more in line with the characteristics of fuzziness and dynamics of music emotion.Experiments show that the analysis model has good music emotion analysis ability.(2)Realize the music recommendation algorithm combined with emotion analysis.Through the application of music emotion analysis model,the music emotion analysis technology is combined with the recommendation algorithm to realize the music recommendation algorithm combined with emotion analysis.The experimental test shows that the algorithm has higher accuracy than the traditional music recommendation algorithm.(3)Through the application of music recommendation algorithm combined with emotion analysis,the music recommendation prototype system based on emotion analysis is realized,including the use description of emotion analysis algorithm and music recommendation algorithm,as well as the demand analysis and implementation of each module function of the recommendation system.The operation results show that the music recommendation algorithm based on emotion analysis has achieved the expected application effect.