MUSIC RECOMMENDATION SYSTEM USING SENTIMENT ANALYSIS

Abstract

Recent studies have already shown that humans respond and react positively to music, and that music has a significant impact on human brain activity. People nowadays frequently prefer to listen to music as a source of entertainment based on their emotions and aspirations. This project concentrates on a system that recommends songs to users based on their emotional state. Computer vision components are used in this system to determine the user's emotion based on facial expressions. Once the emotion is identified, the system recommends a piece of music for that sentimentality, saving the user a significant amount of time over manually selecting and playing songs. It diminishes the time and effort required to manually search for music out of a list based on a person's current state of mind. The CNN algorithm and the Euclidean Distance classifier are used to detect a person's expressions by extracting facial features.