Exploring EEG Frequency Bands as a Feature for Machine Learning in Drowsiness Detection

Drowsiness can be detected by processing brainwaves that was recorded using Electroencephalography (EEG). The EEG signals consist of delta, theta, alpha, beta and gamma frequency bands. This study focuses on EEG frequency bands that were used as a feature in machine learning for drowsiness detection. The feature extractions of those signals were done by using Power Spectral Density (PSD). Several classification algorithms for drowsiness detection, including Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor (KNN), Naïve Bayes, Logistic Regression (LR), Linear Discriminant Analysis (LDA), and Quadratic Discriminant Analysis (QDA) were used in this study. The results indicate that the maximum accuracy achievable is 73.79% when using all frequency bands as a feature with the RF model. However, it should be noted that each model has different characteristics that impact their feature requirements.