A Study on Data Mining Techniques to Improve Student Performance in Primary and Secondary Schools

Authors: Zahira Noor Quraishi, Atul Dattarya Newase

Full-text Research PDF File:   View   |   Download

Abstract: Educational institutions generate large volumes of data related to student demographics, academic records, attendance, and learning behavior. Effectively analyzing this data can support early identification of learning difficulties and improve educational outcomes. This study explores the role of data mining techniques in enhancing student performance at primary and secondary school levels, where timely intervention is most critical. Based on an extensive review of recent Educational Data Mining (EDM) literature, the study examines commonly used methods such as decision trees, random forest, naïve Bayes, artificial neural networks, and feature selection approaches applied to student performance prediction. Existing research highlights that while higher education dominates EDM applications, primary and secondary education remain under-explored despite their long-term impact on learning trajectories. The findings indicate that student behavioral data, academic history, and demographic attributes are strong predictors of performance, and that applying feature selection significantly improves model accuracy. The study emphasizes the importance of data-driven decision-making for educators and policymakers to identify at-risk students and design targeted interventions. By synthesizing current trends, challenges, and gaps, this research provides a foundation for developing effective data mining models aimed at improving learning outcomes and strengthening early-stage education systems.

Keywords: Educational Data Mining, Student Performance Prediction, Primary Education, Secondary Education, Machine Learning

Paper Id: 5.511

Published On: 2024-07-07

Published In: Special Issue - International Conference on Trends and Innovations in Management, Engineering, Sciences and Humanities (July 2024)

Share this