Mental Health Education Evaluation Model Based on Emotion Recognition Algorithm

Abstract

Student learning and development of emotional control abilities are aided by mental health education. It makes it possible for students to understand the sensations and emotions better. Students' mental health and negative emotional responses can help to quickly resolve psychological issues and prevent them from interfering with their regular academic programs. In this study, we proposed a novel Northern Goshwak deep multi-structured Convolutional neural network (NG-DMCNN) to recognize students' emotions related to mental health education. For this study, 300 participants' facial and physiological data were acquired. Pre-processed data using Kalman filtering is an advanced method for data noise reduction. The NG-DMCNN method is compared to the other traditional algorithms. Metrics for performance evaluation include accuracy, precision, recall, and F1-score. The result shows the psychological stress test indicates that the students are in good health and not performing well. The proposed method has superior performance than other algorithms. The study intends to offer a more accurate and effective method for health education.