Smart Learning with Stress Detection: Enhancing Online Education Through AI-Driven Question Answering and Emotional Well-Being Monitoring

The COVID-19 pandemic disrupted traditional classroom learning and paved the way for technology-driven education. Several adaptive personalized learning platforms tailored the education content to individual students’ pace and preferences and monitored students’ performances. One of the significant advances in online learning is the introduction of the Question-Answering (QA) model, which integrates Machine Learning (ML) and Natural Language Processing (NLP) to provide instant, accurate, and context-aware answers to academic queries, reducing students’ dependence on traditional classroom interactions. Although online platforms offered flexibility in learning, they contributed to emotional distress and a higher level of stress. To improve emotional well-being and mental health in online learning, we propose a model called Smart Learning with Stress Detection that enhances the learning experience by integrating Automatic Question Generation (AQG) with a stress detection module that can identify, monitor and respond to student stress levels in real time, providing the necessary interventions before burnout occurs. The novel system incorporates a Text-to-Text Transfer Transformer (T5) and Bidirectional Encoder Representations from Transformers (BERT) to automatically generate and evaluate questions, ensuring comprehensive topic coverage areas. The stress detection module uses Facial Emotion Recognition (FER) to monitor the emotional states of the students while using the platform. The detected emotions are analyzed and recorded in a visual bar chart representation, providing insight into student stress levels. In addition, students can seek guidance from mentors based on their emotional feedback.