An Objective-Based Entropy Approach for Interpretable Decision Tree Models in Support of Human Resource Management: The Case of Absenteeism at Work

摘要: The negative impact of absenteeism on organizations’ productivity and profitability is well established. To decrease absenteeism, it imperative to understand its underlying causes identify susceptible employee subgroups. Most research studies apply hypotheses testing regression models features that are correlated with absenteeism—typically, these limited finding simple correlations. We illustrate the use interpretable classification algorithms for uncovering subgroups employees common characteristics a similar level absenteeism. This process may assist human resource managers in understanding reasons which, turn, could stimulate measures it. Our proposed methodology makes an objective-based information gain measure conjunction ordinal CART model. results indicate model outperforms conventional classifiers and, more importantly, identifies patterns data have not been revealed by other models. demonstrate importance interpretability management through three examples. main contributions this (1) development information-based classifier published dataset (2) illustration approach be considerable value supporting decision-making.