<![CDATA[Attendance is quantitative data that reflects the level of employee discipline and has the potential to be used as a basis for performance evaluation. However, in many agencies, attendance data has not been optimally utilized in making decisions related to employee management. This study aims to classify employee performance based on attendance data using the K-Means Clustering algorithm with the Elbow Method approach. The data analyzed consisted of 57 employees with attributes of the number of tardiness, absence, and blank absences. The research process starts from data collection, pre-processing, to clustering process. Determination of the optimal number of clusters is done with the Elbow Method using Python, which shows the best K value is four. Clustering was done using RapidMiner software, by applying a visual-based workflow to group employees based on similarity in attendance patterns. The final results show the formation of four clusters with different Key Performance Index (KPI) characteristics: excellent, good, fair, and poor. These clusters reflect the level of employee discipline that can be used as a basis for objective managerial evaluation and decision-making. This research shows that the application of data mining-based clustering methods can be a tool in analyzing employee performance in an organizational environment.]]>
Copyrights © 2026
