The Use of Clustering Techniques to Enhance the Accuracy of Islamic Religious Knowledge Assessment

This study aims to enhance the accuracy of Islamic religious knowledge assessment among Madrasah students in Kolaka, Indonesia, by applying the K-Means clustering technique. The background of this research is based on the variation in students' understanding of two important aspects: Fiqh and Aqidah Akhlak. The method used includes data collection from 181 students, followed by descriptive analysis and determination of the number of clusters using the Elbow Method. The results show that the optimal number of clusters is three, representing the categories of High, Medium, and Low. After applying K-Means, a Silhouette Coefficient score of 0.548 was obtained, indicating that the student clustering was effective and valid. These findings provide insights for educators to design learning strategies that are more aligned with the students' needs and highlight the importance of using data analysis techniques in education. Thus, this study contributes to the development of more accurate assessment methods in the context of religious education