The article “Comparative Analysis of Machine Learning Approaches for Drum Set Component Classification in Music Technology” by our head of laboratory Assist. Prof. Dr. Ulaş YURTSEVER and Assist. Prof. Dr. İsmet Emre YÜCEL of the Department of Turkish Music at Amasya University was published open access in Applied Artificial Intelligence on November 7, 2025.
The study classifies the kick, snare, hi-hat, tom, cymbal, and ride sounds of a drum set using spectral and temporal features extracted from audio recordings. Recursive feature elimination with cross-validation (RFECV) combined with gradient boosting was used for feature selection, and support vector machines, random forest, k-nearest neighbors, and decision tree algorithms were then compared. The random forest algorithm achieved the highest accuracy, at 97.1%.
The feature dataset used in the study is also openly available on Zenodo under a CC BY 4.0 license.
The article is available at doi:10.1080/08839514.2025.2581352, and the dataset can be found on our Datasets page.
Etiketler :