Categorization of Grade 9 Students’ Achievement Based on School Examination Results and Qur’anic Achievement Using K-Means and K-Nearest Neighbors

Abstract

Current achievement assessment in Integrated Islamic Schools primarily emphasizes academic performance, with limited recognition of spiritual achievement, thereby constraining comprehensive student evaluation. Therefore, this study proposes a student achievement categorization framework that integrates School Examination Scores and Qur’anic Achievement using a combination of K-Means and K-Nearest Neighbor (kNN) algorithms. This research employed a quantitative approach, including educational data mining, to collect data from 324 ninth-grade students at SMPIT during the 2020–2025 academic years. Data were obtained from academic records and data extraction instruments. The K-Means algorithm in educational data mining served as the primary method for clustering students based on academic and spiritual characteristics, while the resulting clusters were used as target classes for the kNN classification algorithm. Model performance was evaluated using a confusion matrix based on accuracy, precision, and recall metrics. The findings demonstrate that K-Means successfully generated distinct achievement clusters reflecting the diversity of student performance, while the kNN model achieved high performance in consistently predicting the categories of Passed and Conditionally Passed. These results indicate that the integrated K-Means and kNN model is effective in establishing data-driven student achievement categories. From a broader perspective, this study contributes to helping Integrated Islamic Schools develop a student achievement evaluation model that encompasses both academic and spiritual dimensions.