Analysis of Malang University Student Achievement Grouping Using the K-Means Clustering Method

With the increasing number of students and variations in achievement, managing achievement data in higher education has become more complex, so manual methods are insufficient. K-means clustering was chosen because of its ability to group data based on specific attributes, which makes it easier to identify patterns and trends. This research aims to prove K-Means' effectiveness in analyzing achievement data and adding to the literature regarding the application of data mining in education. The dataset includes student achievement indexes from various study programs at the University of Malang from 2018 to 2022. The data is processed to group student achievements efficiently. The clustering model was built using one of the algorithms in the clustering method, namely K-Means. This research produced the best cluster with a total of 3 clusters. The process was conducted to determine the best grouping by testing six cluster models. The best cluster was selected using the Davies Bouldin index test. Based on research with the results, these three groups can be categorized as cluster 0 in the low category with a value of 100, cluster 1 in the high category with a value of 4.100, and cluster 2 in the middle category with a value of 1.900.