AI-Based Pattern Recognition Model for Enhancing Student Engagement and Supporting Academic Planning

With the increase in online education, maintaining student engagement and real-time monitoring has become a significant challenge. Manual monitoring is labour-intensive and ineffective in dynamic virtual learning environments. The purpose of this research is to develop an artificial intelligence (AI)-based pattern recognition model to improve student engagement tracking and academic planning utilizing the PPDAC (Problem, Plan, Data, Analysis, Conclusion) framework. The problem is ineffective engagement monitoring in online situations. The plan entails incorporating AI for emotion and behavior analysis. The data consists of facial expressions and student activity. The analysis uses deep learning (DL) to interpret engagement, and the conclusion supports adaptive, tailored training. Data are collected from Student Engagement and Emotion Recognition Dataset. Use Gaussian blur and Histogram Equalization (HE) to smooth facial images, improve image contrast, and help in low-light situations. Texture-based features are used to recognize expressions using Local Binary Patterns (LBP). This research proposed a novel 3D Convo Flying fox Neural Network (3D-CFNN) model for student engagement monitoring and academic planning. A hybrid proposed model combining a 3D Convolutional Neural Network (3D CNN) for emotion recognition and Flying Fox Optimization (FFO) for parameter tuning was employed. The proposed 3D-CFNN method achieved higher accuracy (0.995), precision (0.988), recall (0.980), and F1-score (0.994) in student identification and emotional state classification, outperforming conventional methods. The proposed AI-based pattern recognition model enables automated, real-time engagement tracking and supports personalized academic planning, leading to improved learning outcomes in virtual environments.