Description of Invention
Music deeply affects emotions and well-being, yet traditional recommendation systems often miss a user’s current mood. This study introduces a Sentiment-Based Music Recommendation System that combines sentiment analysis and machine learning to offer emotion-driven song suggestions. Audio features from platforms like Spotify are analyzed using NLP to classify emotions as positive, negative, or neutral. Sentiment scores, tempo, genre, and other features guide a model to generate mood-aligned recommendations. Built with Python, TensorFlow, and Scikit-learn, the system offers a web interface. It achieved 87.5% sentiment classification accuracy, with 78% of users favoring its emotional relevance over traditional recommendation systems.