Real-Time Student Engagement Prediction in E-Learning Platforms Using Recurrent Neural Networks and Django

Santhosh Kanna J, Dr. B. Kalpana

Abstract: Learning outcomes and academic achievement are significantly impacted by student engagement. Traditional engagement measurement techniques, such surveys and in-person observations, frequently lack scalability and objectivity. In order to automatically evaluate student engagement levels using temporal data, this study suggests a predictive model based on recurrent neural networks (RNNs). The RNN model is fed interaction logs from online learning platforms, including information on how often users check in, how long sessions last, and how they access content. To assign pupils to various involvement groups, the algorithm learns behavioral patterns. By giving teachers timely insights, this strategy seeks to enable early interventions for kids exhibiting low engagement.

Keywords: Student engagement, Recurrent Neural Network, online learning, educational analytics, deep learning

How to Cite?: Santhosh Kanna J, Dr. B. Kalpana, "Real-Time Student Engagement Prediction in E-Learning Platforms Using Recurrent Neural Networks and Django", Volume 14 Issue 8, August 2025, International Journal of Science and Research (IJSR), Pages: 144-147, https://www.ijsr.net/getabstract.php?paperid=SR25802064553, DOI: https://dx.doi.org/10.21275/SR25802064553