Transformer-Based Predictive Model for Strategic Human Resource Management: Applications in Performance, Recruitment, and Retention

Human Resource Management (HRM) plays a vital role within organizations, encompassing a range of functions aimed at maximizing employee performance while aligning with overall strategic objectives. The expansion of global business operations poses significant challenges for human resource management (HRM), including recruiting qualified candidates, retaining talent, and providing adequate training. To address these issues, this study employs Transformer-Based Predictive model which can effectively capture complex patterns in sequential and textual HR data to enhance predictive analytics in HRM processes. In this study, we utilize the IBM HR Analytics Employee Attrition & Performance dataset, to evaluate the effectiveness of our transformer-based method. The performance of our approach is compared against existing baseline methods such as RNN, LSTM, BiLSTM and CNN to demonstrate improvements in predicting employee performance, recruitment outcomes, and retention. The proposed transformer-based approach surpasses traditional methods by providing a deeper understanding of employee behavior and strengthening decision-making processes within human resource management. This study emphasizes the strategic value of advanced analytics in HRM, illustrating how data-driven insights contribute to cultivating a motivated and high-performing workforce, thereby enabling organizations to maintain a competitive advantage in dynamic business environments.

کلیدواژه ها:

Human Resource Management ، Transformer Networks ، Deep Learning ، Recruitment Optimization ، Employee Performance Forecasting