<![CDATA[Student dropout (resign) is a problem that needs to be addressed as early as possible. The number of students dropping out will decrease the quality of the performance of a university, as well as reduce it as much as possible because it will have an impact on public appreciation. As a first step to reducing it, it requires the clustering of students who experience this. Based on this cluster, a pattern of student tendency to drop out can be identified. The parameters used in this study were the GPA, the study period, the number of credits received, and the number of semesters inactive. To compile a cluster, the k-Medoids algorithm is used with 3 types of clusters. Based on the results of the clustering, it can be seen that the dominance of dropout students is due to GPA <2.00 as much as 38.2% and due to not being active in college as much as 52.2%. To measure the cluster quality, the Silhouette coefficient algorithm is used and the resulting coefficient value is 0.3, meaning that the cluster separation rate weak structure.]]>
Copyrights © 2022
