Artificial Neural Network-based Emotion Classification Using Physiological Signal

In this work, we propose an artificial neural network based emotion classification (ANN-EC) with two opposite emotions, happiness and sadness. Compared with the support vector machine based emotion classification (SVM-EC), ANN-EC performs more flexible in classification. Also, ANN-EC shows higher accuracy than SVM-EC. ANN-EC with multiple features reaches 96.25% accuracy and it performs better than just using asingle feature.