Predicative analysis of student performance in online learning

Authors

  • S.V. Durga Prasad Author
  • R. Lahari Author
  • P.Naga Sai Author
  • M. Pavana Sai Author
  • P. Harshavardhan Author

Keywords:

student engagement, binary classification, Random Forest, machine learning, educational technology, digital education, adaptive learning

Abstract

Student engagement plays a vital role in ensuring effective learning outcomes, especially in digital education settings. This study presents a machine learning approach to classify engagement levels as "engaged" or "not engaged" using the Student Engagement Level-Binary dataset. The Random Forest algorithm was utilized, achieving an exceptional accuracy of 100%, showcasing its ability to identify patterns and deliver highly reliable predictions. The trained model has been stored as “student engagement model.pkl”, enabling seamless deployment in real-time educational applications, such as engagement monitoring and adaptive learning platforms. These findings demonstrate the value of leveraging data-driven techniques to support personalized and timely interventions for improved learning experiences. Future research will focus on validating the model with larger datasets and exploring its integration into scalable, real-world systems.

Downloads

Download data is not yet available.

Downloads

Published

15-04-2025

How to Cite

Predicative analysis of student performance in online learning. (2025). International Journal of Information Technology and Computer Engineering, 13(2), 200-207. https://ijitce.org/index.php/ijitce/article/view/1028