Predicting Effect of Study Time and Educational Support on Academic Success and Failure through Machine Learning

Using machine learning, the present study explored the potential to predict student success or failure based on two key factors: the amount of time students dedicate to their studies and the type of educational support they receive. By analyzing a dataset containing information on students' study habits and support systems, the researchers developed a predictive model capable of identifying students at risk of academic difficulties. Amongst the utilized classifiers, Linear Discriminant Analysis (LDA) predicted academic grades with a 91.1% accuracy and academic failure with an 86% accuracy. The prediction models for both success and failure, based on students’ study time and educational support, demonstrated promising results, suggesting that machine learning can be a valuable tool in identifying and addressing students' needs early on, leading to improved academic outcomes. The outputs of such prediction models can help in enhancing learning experiences for students and eventually, help in reducing dropout rates.