Emotional Stability Detection Using Convolutional Neural Networks

Abstract

Emotion recognition is the process of identifying human emotions. It is made possible by processing various modalities including facial expressions, speech signals, biometric signals, etc. Facial Emotion Recognition (FER) has been a growing field since the first works on FER by Ekman in 1970s where he adopted and improved the Facial Action Coding System (FACS). In human-computer interaction, FER is important for several applications in which the user's emotional state is required. The recent years witnessed hugbe advancements in artificial intelligence, specially neural networks; this paper uses convolutional neural network for FER to detect Emotional Stability. We achieve an accuracy of 81% on the classification of neutral, negative and positive emotions.