A Novel Exploitative and Explorative GWO-SVM Algorithm for Smart Emotion Recognition

Emotion recognition or detection is broadly utilized in patient–doctor interactions for diseases, such as schizophrenia and autism the most typical techniques are speech facial recognition. However, features extracted from these behavior-based emotion recognitions not reliable since humans can disguise their emotions. Recording voices tracking expressions a long term also efficient. Therefore, our aim to find efficient scheme, which be used nonbehavior-based real time. This solved by implementing single-channel electrocardiogram (ECG)-based scheme lightweight embedded system. existing schemes have relatively low accuracy. For instance, accuracy about 82.78% using least squares support vector machine (SVM). we propose scheme—exploitative explorative gray wolf optimizer-based SVM (X-GWO-SVM) ECG-based Two data sets, one raw self-collected iRealcare set, widely benchmark WESAD set X-GWO-SVM algorithm Leave-single-subject-out cross-validation yields mean of 93.37% 95.93% set. work demonstrates that exhibits superior performance reliability compared use other supervised learning methods earlier works. It implemented system, much more than solutions based on deep neural networks.