Predicting Absenteeism at Work Using Machine Learning Algorithms

Print Friendly, PDF & Email

Muthanna Journal of Pure Sciences – MJPS

VOL.(7), NO.(1), 2020

Al Muthanna University, Almuthanna, Alsamawa, Iraq

*Corresponding Author : samir@mu.edu.iq

DOI: 10.18081/2226-3284/16-12/1-12

Abstract

To work in the commercial environment, the company needs to be a major competitor in the
business market, which depends mainly on the company’s resources. One of the most important
resources is the employees. Based on that, the absence of the employees from work leads to
deterioration and reduce production in the institutions which leads to heavy losses. There are many
reasons why employees are absent from work. Those may include health problems and social
occasions. The purpose of this paper was to apply machine learning techniques to predict the
absenteeism at work. There are four methods have been used in this research ( neural network(NN)
technique ,decision tree (DT) technique, support vector machine (SVM) technique and logistic
regression (LR) technique. . decision tree model has the highest accuracy equals to 83.33{b68974fa43f6bc28f28a851bd498b9a747a0be965aa6c6cfa2ae6fab94444ced} with AUC
0.834 and the support vector machine has the lowest accuracy equals to 68.47 {b68974fa43f6bc28f28a851bd498b9a747a0be965aa6c6cfa2ae6fab94444ced} with AUC 0.760.

Keywords:

Decision Tree; SVM; ANN; Logistic Regression; Predicting Absenteeism at Work


Down load full article/PDF

Leave a Reply

Your email address will not be published. Required fields are marked *