<![CDATA[High employee turnover can negatively affect organizational stability and productivity, making data-driven predictive strategies essential to identify factors influencing employees’ resignation decisions. This study aims to analyze employee data at PT. Top Remit to predict resignation probability using the Random Forest algorithm, which is known for its strong performance in handling complex and heterogeneous data. Using a machine learning approach, this research processes historical employee data that include demographic factors, job satisfaction, compensation, and performance indicators. The dataset consists of labeled employee records based on their employment status. The Random Forest model was applied to identify patterns contributing to employee turnover and is evaluated using performance metrics such as accuracy, precision, recall, and F1-score. The results indicate that the optimized Random Forest model, after hyperparameter tuning using Grid Search, achieves a test accuracy of 82.5%, with a resignation-class precision of 1.00 and a recall value of 0.6316. Furthermore, feature importance analysis reveals that the most influential factors affecting resignation decisions are annual performance scores, average review scores, compensation satisfaction, job satisfaction, and work engagement. Based on these findings, the predictive model can be utilized as a decision-support tool to design more targeted and data-driven employee retention strategies. This study is expected to contribute to the development of predictive analytics in human resource management, particularly in the application of machine learning methods to support workforce management and sustainability.]]>
