Isolation Forest-Based Employee Performance Analysis for Identifying Burnout and Engagement Issues

Organizations have significant challenges due to employee burnout and engagement difficulties, which have a detrimental effect on both productivity and general wellbeing. Conventional approaches frequently fail to adequately detect these problems because of their limitations in terms of precision and dependability. This study aims to tackle these difficulties by introducing an innovative approach that use the Isolation Forest model to identify irregularities in employee behaviour and performance data. The suggested approach aims to precisely detect indications of burnout and engagement issues by identifying abnormal patterns in the dataset, offering a reliable tool for timely intervention. Comparative investigation reveals that the Isolation Forest model surpasses other current technologies, such as Random Forest and XGBoost, with notably superior metrics. The model attains an accuracy of 0.98, precision of 0.97, recall of 0.96, F1-Score of 0.965, and ROC-AUC of 0.99. The results demonstrate the model’s exceptional capacity to differentiate between normal and abnormal behaviour, providing a more efficient method for monitoring and improving employee wellbeing. This study enhances the field by introducing a more precise and dependable approach for identifying burnout and engagement problems, hence promoting improved organizational practices and employee support systems.