<![CDATA[This study aims to implement the Decision Tree C4.5 algorithm in the classification of permanent employee appointments at PT Intinusa Teknik Sejahtera to support more objective, accurate, and transparent decision-making. The research method used is a quantitative data mining-based approach with the CRISP-DM framework, using an employee dataset from the HRD department consisting of 16 attributes related to profile and performance. The modeling process was performed using RapidMiner Studio software using the split validation method with a ratio of 80% training data and 20% test data. The results show that the Decision Tree C4.5 classification model has an accuracy of 89.25%, with recall for the Contract class of 91.14% and the Permanent class of 78.57%, and precision for the Contract class of 96.00% and the Permanent class of 61.11%. The conclusions of this study confirm that the attributes of Performance, Attendance, Loyalty, and Tenure are the main factors in permanent employee appointments, and the C4.5 algorithm can be utilized as an HRD decision support system, although further method development is needed to improve precision for the Permanent class. Keywords: Decision Tree C4.5, Classification, Permanent Employees, Human Resource Management]]>
Copyrights © 2025
