<![CDATA[This study explores the application of the Naive Bayes classification method to predict student grades based on important attributes such as timeliness of assignment submission, attendance rate, and quality of work. This research uses a dataset that includes three attributes, namely timeliness of submission, attendance level in learning, and evaluation of the quality of assignments collected by students. The pre-processing is performed to clean the data, followed by an under-sampling stage to balance the class distribution. Then, the classification model is evaluated and tested using specific data samples to measure prediction accuracy. The results showed a significant improvement in model accuracy after applying under-sampling, highlighting the importance of handling data imbalance in predictive analysis. The implications of these findings are not only relevant in the context of higher education, but also offer opportunities for further development in data-driven decision-making in various fields.]]>
Copyrights © 2025
