Enhancing Teacher Support Systems with Emotion-Aware Learning Using Affective Computing and Gans

Abstract

The psychological state of teachers is an essential but relatively understudied component of the quality of education. Stress, emotional exhaustion, and burnout of teachers have adverse consequences on the performance of students and overall productivity within organizations, as well as higher attrition rates of educators. Although there is an increasing understanding of the issues facing today's educators, contemporary educational assistance systems fail to consider affective monitoring and adaptation during interactions. This paper presents a new approach for addressing the problem of psychological wellness of teachers, the Emotion-Aware Teacher Support System (EA-TSS). The presented framework combines Affective Computing methods with a Generative Adversarial Network (GAN) architecture to provide personalized real-time support to educators based on their emotions. In the suggested framework for EA-TSS, multimodal affective sensing (such as facial action units, voice prosody analysis, and physiological measurements) is used. The emotion classification task is performed by an attention-based BiLSTM model, which attains an accuracy of 94.7% on the AffectNet-Teacher dataset containing 8,400 labeled samples with seven distinct emotion classes. The generated content is tested through FID score, with FID = 18.3 establishing perceptual fidelity. Empirical findings reveal that EA-TSS lowers the teachers' reported burnout index by 31.4% during a 12-week implementation period, showing a statistically significant increase in the perceived quality of the intervention
(p < 0.001). Benchmarking against five competing baselines affirms the superiority of the proposed solution in terms of Accuracy (94.7%), F1-Score (93.2%), AUC-ROC (0.971), MAE (0.083), and Recall (92.8%). Ablation analysis corroborates the importance of all architectural modules. The novel approach makes a valuable contribution to the domain of affective computing in education technology.