1. A system for recommending music based on emotions.
- Author
-
Sravanthi, T., Zuya, Razeena, Samreen, Sabahath, Basri, Rabia, Rithima, P., and Pasha, Syed Nawaz
- Subjects
MUSIC & emotions ,MUSICAL perception ,GALVANIC skin response ,EMOTION recognition ,SUPPORT vector machines ,MUSICAL aesthetics - Abstract
The majority of currently used music recommendation systems rely on content-or collaborative-based recommendation engines. However, a user's choice of music is not only based on past musical tastes or its actual substance. but also based on how that individual is feeling. This study suggests a framework for emotion-based music recommendations that can identify a user's mood based on data from wearable physiological sensors. A wearable computing device that includes physiological sensors for galvanic skin response (GSR) and photo plethysmography (PPG) specifically categorises a user's emotion. The emotion data is included as additional data to any collaborative or content-based recommendation engine. These data can therefore be used to enhance the functionality of the current recommendation engines. Since arousal and valence can be predicted from a variety of physiological variables, the problem of emotion recognition in this work is seen as such. On the GSR and PPG signal data of 32 people, with or without feature fusion, experimental results are obtained using decision tree, random forest, support vector machine, and k-nearest algorithms. The results of in-depth experiments using real data show that the suggested emotion classification method, which can be used, is effective. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF