<![CDATA[Technological developments occur so quickly, resulting in continuous changes that qualified human resources are needed to support the endless times that run. This study will classify individuals with information technology and computer skills in Indonesia based on region. This research used K-Means clustering, the Rank Order Centroid method, and the Davies-Bouldin Index clustering evaluation method to assess accuracy. K-means clustering is a simple algorithm and does not require a target class. There are areas for improvement in the K-Means process, namely at the initial centroid determination stage. Therefore, the ROC method is used. Based on data taken from the website of Badan Pusat Statistik Nasional about the proportion of productive age individuals 15-59 years who have Information and Computer Technology skills by the province during 2017-2021. It produces 3 clusters, including a high-level cluster in which there are 8 provinces, a medium-level cluster in which there are 22 provinces, and a low-level cluster in which there are 4 provinces, and obtained a DBI value of 0.163625 which is close to 0, meaning that the quality of the accuracy of the clustering results is good. Based on clustering results with good accuracy, using K-Means can be combined with ROC and is quite effective. The government can use the results of this study to prioritize improving the quality of human resources in areas with low-level information and computer technology skills. Suggestions for further research using other clustering algorithms and ROC as a comparison.]]>
Copyrights © 2023
