Predictive Analytics for Human Resource Management: Enhancing Talent Acquisition and Retention Using Data analytics

HRM depends on the output of predictive analytics by which the input achieves accurate data on employee hiring and retention. For enhancing the decision-making in the HR domain, this research proposes a novel multi-modal logistic regression, XGBoost, and deep learning-based designated hybrid prediction model. Due to the integration of several more common and uncommon HR datasets and richer feature engineering techniques, the mentioned model eliminates the limitations of previous studies. These enhance the chances of applicant selection and consequently the chances of the applicants' retention. The proposed hybrid model achieved 91.4% accuracy for applicant selection and 89.7% for employee retention; the accuracy of the models developed in the years 2021, 2022 & 2023, was 82.3%, 84.9%, and 87.4% respectively. Its implementation using IBM Watson and Kafka will afford HR managers practical real-time information from the model. Lastly, the benefits of applying the model to enhance the HR procedures are highlighted and recommendations for further work to increase the model's applicability in other areas are made.