Using Data Mining Techniques to Predict Workforce Needs: A Literature Review

This paper aims to review various studies that use data mining techniques to estimate the supply and demand for labor in various industries. To investigate the data mining models used to predict the size of the workforce and the accuracy given in this study. After reviewing the literature to evaluate the prediction approaches, these twelve publications met the study's requirements and selected suitable methodologies. As a result, this study illustrated that various machine learning algorithms were used among the studies reviewed in the field of forecasting personnel recruitment, such as neural network algorithms, XGBoost algorithms, regression models, Gaussian process regression (SVM), Random Forest regression, and many others. It also turns out that no model is better than the others in terms of performance. The most widely used algorithms are Artificial neural networks (ANN), support vector machines (SVM), and XGBoost, which slightly outperformed the rest of the algorithms. According to the review of these studies, the factors affecting workforce supply and demand depend on the field's circumstances. This study investigates how the chosen studies address the labor shortage issue using machine learning techniques.