Emotion-driven music recommendations: Integrating CNN and KNN for personalized playlists

No Thumbnail Available
Gikonyo Kamau, Nelson
Issue Date
Masters in Artificial Intelligence
Dublin Business School
Items in eSource are protected by copyright. Previously published items are made available in accordance with the copyright policy of the publisher/copyright holder.
This thesis delves into the captivating world of personalized music playlists, where emotions, technology, and user preferences converge. Through a unique blend of sentiment analysis, Convolutional Neural Networks (CNN), and K-Nearest Neighbors (KNN), we embark on a journey to create playlists that resonate with your feelings. The CNN deciphers emotions from facial expressions, shaping the emotional landscape, while KNN fine-tunes song recommendations for a harmonious experience. To ensure accuracy, a gradient boosting model steps in to validate emotional connections. We also explore the power of user feedback loops and the potential of multi-modal emotion recognition. With a strong ethical compass and an interdisciplinary approach, we uncover the profound connection between emotions and music recommendation. The results magnify the importance of emotions in shaping musical experiences, leading to a symphony of personalized playlists