Monitoring Students in Online Learning Environments Using Deep Learning Approach

Since education is rapidly growing digital, it is more vital than ever to keep a check on students' online activity to make sure they are learning properly. The highly effective Convolutional Neural Network model introduced in this paper is specially developed to recognize student distractions in online learning settings with high precision and minimum processing cost. Our approach is perfect for real-time applications on low-processing-power computers since it uses fewer than 800,000 parameters and achieves an impressive accuracy of 96.13%, in contrast to classic models with many parameters. In order to detect subtle behavioral markers of student engagement and distraction without requiring a lot of computing resources, the proposed method uses a CNN architecture. The model's capacity to generate accurate predictions with low latency was validated through a series of tests conducted in online learning settings with a variety of student behaviors. Our study demonstrates that high-performance engagement monitoring is possible with reduced model complexity, adding a valuable tool for online platforms and educational organizations to enable focused, effective learning experiences. The proposed approach paves the way for future advancements in low-cost, highly accurate student monitoring systems and allows for further research into flexible learning environments.