Pamukkale Univ Muh Bilim Derg. Ahead of Print: PAJES-59321 | DOI: 10.5505/pajes.2024.59321

Performance Evaluation of the Machine Learning Algorithms for Emotion Classification on the CASE Dataset

Emre Rifat Yýldýz, Yýltan Bitirim
Department Of Computer Engineering, Faculty Of Engineering, Eastern Mediterranean University, TRNC

Emotion classification using physiological signals is still a challenging task even the sensor technology and machine learning algorithms evolved within the decades. In this study, the performance of KNN, DT, RF, LR, and XGB algorithms on emotion classification was evaluated in terms of accuracy on the CASE dataset. Three sub-datasets namely Downsampled, Resampled-EM, and Resampled-VA were obtained from the original dataset. Then, hyperparameter tuning was applied to the smallest dataset and the algorithms were applied with the parameters that were obtained in hyperparameter tuning to the Resampled-EM, Resampled-VA, and original sets. As the results obtained, KNN, RF, and XGB provided higher accuracies on the Resampled-VA set when compared to the Resampled-EM set, where it was the contrary for the DT algorithm. XGB algorithm provided the highest accuracy of 97.44% among all the results. This study could be considered as the most comprehensive study that utilizes machine learning algorithms for emotion classification on the CASE dataset.

Keywords: Emotion recognition, machine learning, physiological signals, CASE dataset

---

CASE Veriseti Üzerinde Duygu Sýnýflandýrma Ýçin Makine Öðrenmesi Algoritmalarýnýn Performans Deðerlendirmesi

Emre Rifat Yýldýz, Yýltan Bitirim
Bilgisayar Mühendisliði Bölümü, Mühendislik Fakültesi, Doðu Akdeniz Üniversitesi, KKTC

Sensör teknolojisi ve makine öðrenimi algoritmalarý birkaç on yýl içinde evrimleþmiþ olsa da fizyolojik sinyalleri kullanarak duygu sýnýflandýrmasý hala zorlu bir görevdir. Bu çalýþmada, KNN, DT, RF, LR ve XGB algoritmalarýnýn CASE veriseti üzerinde duygu sýnýflandýrmasýndaki performanslarý deðerlendirildi. Orijinal verisetinden Downsampled, Resampled-EM ve Resampled-VA olarak isimlendirilen 3 alt-veriseti elde edildi. Daha sonra, en küçük boyuta sahip verisetine hiper parametre ayarlamasý uygulandý ve algoritmalar hiperparametre ayarlamasýnda elde edilen parametrelerle Resampled-EM, Resampled-VA ve orijinal setlere uygulandý. Elde edilen sonuçlara göre, KNN, RF ve XGB algoritmalarý Resampled-VA setinde DT algoritmasýna kýyasla daha yüksek doðruluklar saðladý. Bu durum Resampled-EM seti için tam tersi olarak gözlemlendi. XGB algoritmasý, %97.44 ile tüm sonuçlar arasýnda en yüksek doðruluðu saðladý. Bu çalýþma, CASE verisetinde duygu sýnýflandýrmasý için makine öðrenimi algoritmalarýný en kapsamlý þekilde kullanan çalýþma olarak deðerlendirilebilir.

Anahtar Kelimeler: Duygu tanýma, makine öðrenmesi, fizyolojik sinyaller, CASE veriseti

---

CITE