The Use of Data Mining Techniques to Predict Employee Performance: A Literature Review

Human resources management considers performance appraisal a critical process, regardless of the organization’s size and level. organizations must compare the performance of their employees with what is expected as well as take precautionary measures and corrective measures to avoid a decline in their performance. In order to address this issue, prediction techniques are used to help determine which individual skills are appropriate for the job. This paper presents a brief overview of the data mining techniques used to predict employee performance and accuracy. A literature review was conducted to evaluate prediction techniques, which resulted in the selection of ten studies that met our criteria and had acceptable methodology and information. The best algorithms used in these studies were RanKer (a novel algorithm) and XGBoost, with an accuracy of 96.25% and 95.S3%, respectively. A review of these studies found that current on-the-job training, motivation, and experience were the most influential factors affecting performance. This study lays the groundwork for human potential prediction work, which may serve as a starting point for developing automated systems in which employee prediction is considered a continuous value rather than a prediction of the employee’s performance category.