AI-Based Predictive Analytics for Student Academic Performance

Abstract

Educational institutions generate large volumes of academic and behavioral data through Learning Management Systems (LMS), attendance portals, and assessment platforms. Leveraging this data using Artificial Intelligence (AI) can significantly improve academic planning and student outcomes. This research proposes an AI-based predictive analytics framework to forecast student academic performance using machine learning techniques. The model analyzes historical academic records, attendance, assignment submissions, and engagement metrics to predict student performance categories (Excellent, Average, At-Risk). Experimental evaluation demonstrates that ensemble-based models outperform traditional classifiers in accuracy and reliability. The proposed system can assist educators in early identification of at-risk students and enable timely academic interventions.

Introduction

Artificial Intelligence is increasingly being adopted in education to analyze the large volumes of data generated by Learning Management Systems (LMS). Despite this, much of the data remains underutilized for proactive academic decision-making. Predictive analytics offers a solution by using machine learning techniques to forecast student performance, identify learning difficulties early, and support timely interventions.

This study proposes an AI-based predictive analytics model to assess student academic performance using academic and behavioral data such as attendance, internal assessments, assignment scores, and LMS activity. A review of existing literature highlights the evolution from simple models like Decision Trees and Naïve Bayes to more advanced ensemble and deep learning approaches, while also noting gaps in scalability and practical implementation.

To address these challenges, the research develops and evaluates multiple machine learning models—Logistic Regression, Decision Tree, Support Vector Machine (SVM), and Random Forest—using a synthetic dataset of 300 students. After preprocessing and feature encoding, students are classified into performance categories: Excellent, Average, and At-Risk. Experimental results show that the Random Forest model performs best, achieving an accuracy of 89.3%, outperforming other models across precision, recall, and F1-score.

The analysis identifies attendance percentage and internal assessment marks as the most influential predictors of academic success. Overall, the proposed system demonstrates the effectiveness of AI-driven predictive analytics in early identification of at-risk students, enabling timely academic interventions and improved educational outcomes.

Conclusion

This study demonstrates the potential of AI-based predictive analytics in forecasting student academic performance. The proposed framework enables early identification of at-risk students and supports data-driven academic interventions. Future work may include integrating deep learning models and real-time analytics within LMS platforms.

References

[1] Romero, C., & Ventura, S. (2020). Educational Data Mining and Learning Analytics: An Updated Survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery. [2] Baker, R. S., & Inventado, P. S. (2019). Educational Data Mining and Learning Analytics. Springer. [3] Han, J., Kamber, M., & Pei, J. (2021). Data Mining: Concepts and Techniques. Morgan Kaufmann.

Copyright

Copyright © 2026 Mr. Juned Juber Shaikh, Baig Ishrat Shamsher, Affan Asif Khan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.