An Ensemble Learning Approach for Automatic Emotion Classification of Sri Lankan Folk Music

Music experience is closely associated with our moods and emotions. Even though data mining techniques have been widely adopted in computational analysis of music-emotion, traditional music including Sri Lankan folk music is less explored computationally. Therefore, considering a Sri Lankan folk music dataset, performed the classification using Support Vector Machines, Naive Bayes, Random Forest (RF), k-Nearest Neighbor (k-NN), and Logistic Regression (LR), employing dynamics, rhythm, timbre, pitch, and tonality features. k-NN achieved the maximum accuracy (78.44%) while RF and LR achieved accuracies of 76.19% and 73.42%, respectively. Combining the above three classifiers, an ensemble model was developed. Max-voting was applied, and the results were further enhanced using ensemble boosting. With optimized features, AdaBoost (RF as base estimator) achieved the highest accuracy (95.23%) while reducing the training time significantly. Expanding the dataset in terms of the number of music stimuli and emotion categories looked progressive.