Machine Learning Analysis on Students’ Demographic and Performance to Predict On-Time Graduation: A Case Study in Indonesia

Achieving on-time graduation is a crucial milestone in the educational journey, reflecting not only individual success but also the effectiveness of educational institutions in providing necessary support. This work-in-progress study proposes a machine learning approach integrating demographic features and academic metrics to predict on-time graduation, a vital educational milestone. Utilizing a dataset from an Indonesian university, we employ supervised and unsupervised learning models. Our analysis assesses model performance, identifies significant predictors, and develops a dynamic warning system. Our analysis anticipates identifying demographic factors contributing to on-time graduation, predicting graduation likelihood using supervised learning, and developing a dynamic warning system for timely interventions. The study’s findings will guide tailored interventions and advance educational data analytics, offering contextual insights into the Indonesian educational landscape.