Smart Song Recommendation System using Machine Learning

The smart music recommendation system has a lot of research scope, and many renowned researchers are working on the recommendation system. Machine Learning is the backbone of almost all the recommendation systems available in the literature. Till now, most of the recommendation system available in the past literature, works on either users’ past preference or watched list or the type of manual preferences the user sets. In this paper, a unique approach to developing a recommendation system is proposed that takes into account the user’s current mood. This system utilizes real-time video/image capturing capabilities through a camera or webcam to detect the user’s current emotions. A neural network-based approach is used. In this paper to design an intelligent music recommendation system a CNN model helped to increase the efficiency level of the overall designed system, and the results obtained are more accurate. For clustering songs, the unsupervised learning K-Means Algorithm is used. The songs’ order will be based on the current popularity of the songs. The proposed design is validated with the help of various simulations.