Abstract - In this project, the prospect of design, implemented and analyzed song recommendation system. The scope of this project to Make a system in that using face recognition songs are recommend. On the basis of face expression the user justify their mood and then recommend the song. Users used Million Songs in Dataset. In this system Kaggle dataset is used. For the correlations between users and songs various dataset used. This system learns previous listening history of users to provide recommendations for songs which users would prefer to listen most and also learn their emotions. Collaborative filtering is the technology that is used for the relationships between users and between items to make a prediction.
Key Words: Recommendation systems, Music, Million Song Dataset, Collaborative filtering, content-based, Face Expressions, Emotions.