Prediction of UiTM student academic performance using Naive Bayes algorithm / Muhammad Irfan Zahin Jailani

This paper proposes a predictive system using the Naive Bayes algorithm to solve the pressing issue of low academic performance among students at Universiti Teknologi MARA (UiTM).The study underlines a variety of factors affecting student outcomes while highlighting the significance of effective academic performance. With the help of customized interventions and early identification of at-risk pupils, the proposed approach seeks to increase graduation rates and overall achievement.The main objectives of this study include studying the Naive Bayes algorithm in student academic performance prediction, designing and developing a student academic performance prediction model utilizing Naive Bayes, and evaluating the accuracy of the prediction prototype using the developed model. The approach, which consists of splitting the training and testing datasets, preparing the data, and applying Naive Bayes, produces remarkable results: 94.85% accuracy, 91.52% precision on average, 96.84% recall, and a 94.08% overall F1 score. As a result, the suggested system promotes improved student achievement and the welfare of society by providing a proactive way of addressing academic difficulties.