Modeling Of Student Graduation Prediction Using the Naive Bayes Classifier Algorithm

The background of this study emphasizes the significance of the graduation accuracy rate as a critical metric for evaluating the caliber of colleges and students. Ensuring accurate predictions of student graduation timelines is vital for university administration to maintain consistent graduation rates and uphold the quality of graduates. The objective was to develop a predictive model using the Naive Bayes classifier to forecast student graduation accurately. The method involved applying techniques such as Log-Probability, Handling Rare Data (Sparse Data), Efficient Implementation, and Laplace Smoothing to process the data. The dataset used in this research comprises student records from the year 2023, sourced from universities in Libya. The results indicate that the model successfully predicted that the number of students who passed (175) was significantly higher than those who failed (25), achieving a prediction accuracy of 94%. This high accuracy demonstrates the effectiveness of the Naive Bayes classifier in graduation prediction. The conclusion drawn from this study is that the Naive Bayes classifier approach is highly promising for forecasting student graduation outcomes, providing valuable insights for university administrations to implement timely interventions and improve overall graduation rates.