Enhancing Organizational Efficiency with Machine Learning-Based Human Resource Information System

Human Resource Information Systems (HRIS) are essential for the management of employee data and making HR decisions. The application of ML can enhance HRIS through the efficiency of resource allocation, forecasting outcomes such as employee turnover and performance, and generally enhancing organizational effectiveness. The purpose of this research is to develop a model called Enhanced Golden Jackal Optimization-driven Random Forest (EGJO-RF) to optimize HRIS. Initially, the dataset comprising information about all employees working at a specific company is applied. Then the data was preprocessed using Z-score normalization for the scaling of numerical features, namely salary and performance. As principal component analysis (PCA) reduces dimensions without losing all important information, which helps to diminish dimensionality. The EGJO-RF model considerably improves HRIS by enhancing prediction accuracy, resulting in more effective decision-making and higher employee performance. Experimental result shows that EGJO-RF achieves 99.3% accuracy, 92.7% precision, 78% recall, and 84% F1 score. The EGJO-RF model shows better predictive abilities, especially for high-dimensional HR data, and also enhances accuracy in effectively predicting employee turnover, thus optimizing resource management and HR decision-making.