<![CDATA[The development of information technology has encouraged educational institutions to make more optimal use of academic data in order to improve the quality of learning evaluation. Data such as assignment scores, quizzes, midterm exams, final exams, and student attendance are no longer just administrative archives, but can be processed using Educational Data Mining (EDM) techniques to generate new information that supports the academic decision-making process. This study aims to build a model for predicting students' final grades using the Decision Tree algorithm by utilizing these academic attributes as input variables. The research process was carried out in several stages, starting from data collection, preprocessing, data transformation, classification model formation, to performance evaluation using a confusion matrix and accuracy calculations. The results of the study show that the Decision Tree algorithm is capable of classifying students' final grades with an accuracy of 80%. Feature importance analysis reveals that final exam scores are the most influential attribute in the formation of decision tree structures, followed by midterm exam scores, while assignment, quiz, and attendance scores contribute less. These findings indicate that summative evaluation plays a dominant role in determining students' final grades. Overall, this study proves that Decision Tree is an effective classification method that is easy to interpret and highly relevant for use in the context of EDM, especially in helping schools conduct objective and data-driven student performance analysis.]]>
