Electroencephalogram (EEG) is used to analyze the state of the brain. One of the critical states of the brain is drowsiness. Physical, mental tiredness, and unconsciousness are some of the reasons for drowsiness. Drowsiness state may lead to fatal crashes, severe injury, and property damage; sometimes, it can be analyzed and detected by using EEG. Analyzing EEG signals is complicated and tedious, so an automated diagnosis is required to interpret these signals effectively. In recent years, finding drowsy feeling while working has become an important research area. In this paper, the authors reviewed various drowsiness detection techniques in the literature and analyzed the performance of 15 different machine learning algorithms for the self-acquired feature set from the EEG.

脑电图（EEG）用于分析大脑的状态。大脑的关键状态之一是困倦。身体、精神疲劳和意识不清是嗜睡的一些原因。瞌睡状态可能导致致命车祸、重伤和财产损失；有时，可以通过脑电图来分析和检测。分析脑电图信号既复杂又乏味，因此需要自动诊断来有效地解释这些信号。近年来，寻找工作时的困倦感已成为一个重要的研究领域。在本文中，作者回顾了文献中的各种睡意检测技术，并分析了 15 种不同的机器学习算法针对从脑电图自获取的特征集的性能。