Generative AI-Enabled Human Resource Management for Adaptive Talent Acquisition and Employee Engagement

Abstract

As a growing industry, Human Resources Management (HRM) heavily depends on intelligent technology to grow its talent acquisition and engagement capabilities. However, current methodologies have a number of advantages such as lack of workforce analysis, recruitment & engagement data integration challenges, and lack of consistency in decision-making support. This paper offers the ‘Progressive Workforce Intelligence Generation' (PWIG) framework for enhancing the intelligent HR decision making process. PWIG includes tasks such as data integration, data preprocessing, feature engineering, encoding categorical features, feature selection, feature scaling, data balancing, hyperparameter tuning, and multi-model classification by models like Lightning Gradient Boosting Machine (LightGBM), AdaBoost, linear Support Vector Machine (SVM), and model Gated Recurrent Unit (GRU). Based on those classification results the framework generates intelligent patterns in the workforce, which can be utilized for recruitment and employee engagement decisions. The effectiveness of the framework is tested using the evaluation indicators namely Balanced Error Rate (BER), Matthews Correlation Coefficient (MCC), Cohen's Kappa, Log loss and Decision time. In conclusion, the PWIG approach presents a comprehensive and adaptable solution to HRM problems, enabling better human resource decision making and operational efficiency. The overall precision of the proposed classification system, PWIG was 97.1%.