Automated User Engagement Analysis Through Multi-Modal Facial Expression Recognition: A Deep Learning Approach

Automated user engagement analysis using artificial intelligence (AI) has become increasingly critical across education, digital marketing, and human-computer interaction. Traditional approaches are subjective and lack adaptability in real time. This study proposes a deep learning-based multi-modal framework integrating Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks with dual-attention mechanisms to assess engagement using facial expression recognition. The proposed model achieves 89.3% classification accuracy across seven emotional states and five engagement levels using AffectNet and RAVDESS datasets. The system is robust to environmental variations, performs in real time, and opens future research directions with physiological signal integration and edge deployment.