ARCHIVES
VOL. 10, ISSUE 2 (2025)
Harmonizing emotions: Deep learning approaches for music emotion classification
Authors
Arjun Mathur, Priyanshu Sharma, Shivam Yadav, Dr. Meena Chaudhary, Dr. Gunjan Chandwani
Abstract
Music has a profound impact on human emotions, influencing mood,
relaxation, and motivation. This research paper presents a novel approach to
classifying songs into three distinct emotional categories—meditation,
motivational, and sad—by deriving mathematical formulas from their beats and
tunes. We analyze rhythmic patterns, tempo variations, harmonic structures, and
spectral features to quantify musical characteristics that define each
category. A machine learning model is trained using these features to automate
classification. Our experiments demonstrate that beat intensity, tempo
stability, and harmonic complexity are key discriminative factors, achieving an
overall classification accuracy of 89.3%. This study contributes to the fields
of music information retrieval (MIR) and affective computing by providing a
mathematically grounded framework for emotion-based song classification.
Download
Pages:89-95
How to cite this article:
Arjun Mathur, Priyanshu Sharma, Shivam Yadav, Dr. Meena Chaudhary, Dr. Gunjan Chandwani "Harmonizing emotions: Deep learning approaches for music emotion classification". International Journal of Advanced Education and Research, Vol 10, Issue 2, 2025, Pages 89-95
Download Author Certificate
Please enter the email address corresponding to this article submission to download your certificate.