This review analyzes current trends in using electrocardiogram and electrodermography signals as training features for machine learning classifiers to perform emotion classification.
This paper reviews emotion classification investigations, focusing on the use of the Electrocardiogram (ECG) and Electrodermography (EDG)/Galvanic Skin Response (GSR) as input features. Currently, a large majority of emotion classification studies utilize Electroencephalograms (EEG) and facial expression recognition to perform emotion classification. Fewer studies have been conducted using the ECG and EDG to this end. These physiological signals will be reviewed to compare the ECG and EDG approach, equipment, and stimuli used, as well as machine learning algorithms utilized to perform the classification task. The main objective of this paper is to analyze the current trends in terms of how signals including heart rate and skin conductance can be used as training features for machine learning classifiers to perform the emotion classification task. Some critical observations and open problems will be presented, followed by a discussion of promising avenues for future research in the use of ECG and EDG for emotion classification.
Bulagang et al. (Wed,) conducted a review in Emotion classification. Electrocardiogram (ECG) and Electrodermography (EDG) signals was evaluated. This review analyzes current trends in using electrocardiogram and electrodermography signals as training features for machine learning classifiers to perform emotion classification.