Paper Contents
Abstract
Predicting how well a student might do academically is something educators and institutions are always working on. There are a bunch of factorspersonal, academic, even environmentalthat can influence student performance, and figuring out how they all come together isnt easy. In this project, I used data mining techniques to try and make sense of it. The idea was to analyze past data to see if we can predict students final academic outcomes. I worked with a real-world dataset of undergraduate students, and applied different classification algorithmsDecision Trees, Nave Bayes, and k-Nearest Neighbors (KNN)to see which one worked best. I ran the models using 10-fold cross-validation to make sure the results were reliable. Among all the algorithms, the Decision Tree turned out to give the highest accuracy. So, based on this, I believe data mining can be a really helpful tool in forecasting academic performance, and it could eventually help educators take early action for students who might be struggling.
Copyright
Copyright © 2025 Shubha R. This is an open access article distributed under the Creative Commons Attribution License.
