In order to categorize data from the findings of the MBKM survey using machine learning, the goal of this research is to determine how well the implementation of MBKM has been comprehended by Education staff, lecturers, and students at the university level. Artificial intelligence, which is frequently employed to address a variety of issues, includes machine learning. K-Mean and K-Medoids clustering algorithm models were used to group the data for suggestions on how to apply MBKM at Bhayangkara Jakarta Raya University. This study uses databases and machine learning approaches to categorize MBKM survey data at the level of the relevant study program. In order to examine the machine in accordance with the learning algorithm, the K-Mean and K-Medoids clustering algorithms will be utilized. This research will train a machine or system using filtered data that can predict the outcomes of the MBKM survey, which was constructed using machine learning. The study's findings come from the application of MBKM into two groups, with cluster 1 showing a high degree of comprehension and cluster 2 showing a low level of understanding, as produced by K-Mean. In the meantime, K-Medoids created cluster 2 for high comprehension of MBKM and cluster 1 for low understanding of MBKM implementation. The results of the comparison evaluation of clustering between K-Mean and K-Medoids obtained cluster evaluation values using the Davies Bouldin Index by conducting trials starting from K=2, K=3, K=4 and K=5 showing lower K-Mean values compared to K-Medoids, so that K-Mean is recommended as a clustering algorithm for grouping the results of the MBKM survey implementation in Higher Education.
Copyrights © 2023