Prediction of Absenteeism at Work using Data Mining Techniques

Abstract

High absenteeism among employees can be detrimental to an organization as it can result in productivity and economic loss. This paper looks into a case of absenteeism in a courier company in Brazil. Machine learning techniques have been employed to understand and predict absenteeism. Understanding this would provide human resource managers an excellent decision aid to create policies that can aim to reduce absenteeism. Data has been preprocessed, and several machine learning classification algorithms (such as zeroR, tree-based J48, naive Bayes, and KNN) have been applied. The paper reports models that can predict absenteeism with an accuracy of over 92%. Furthermore, from an initial of 20 attributes, disciplinary failure turns out to be a very prominent feature in predicting absenteeism.