Student graduation is one of the primary indicators used to evaluate the success of the educational process in a school. Graduation rates not only reflect students’ academic achievement but also indicate the effectiveness of learning strategies, the quality of school management, and the success of student development programs. However, schools often encounter difficulties in identifying patterns and factors influencing student graduation due to the large volume of academic data, diverse attributes, and complex relationships among variables. This study aims to apply data mining techniques to analyze student graduation patterns using the Naïve Bayes method, a classification approach based on Bayes’ theorem with the assumption of independence among attributes. The data used in this study include report card scores, school examination scores, attendance records, and student behavior related to graduation status. The data processing was carried out through the Knowledge Discovery in Database (KDD) stages, which include data selection, data cleaning, data transformation, application of the Naïve Bayes algorithm, and evaluation of classification results using a confusion matrix. In the data selection stage, attributes relevant to graduation status were chosen from the school’s academic database, followed by data cleaning to address missing, inconsistent, or duplicate data. The data were then transformed into a suitable format for processing by the Naïve Bayes algorithm. The results of the study indicate that the Naïve Bayes method is capable of classifying student graduation status with a good level of accuracy. Furthermore, academic scores, attendance, and student behavior were found to significantly contribute to the probability of student graduation. Therefore, the Naïve Bayes method can be utilized as a decision-support tool for schools in identifying students at risk of not graduating, designing early intervention programs, and improving academic quality more effectively.
Copyrights © 2025