Decoding the Indian Workforce an open-Source Analytics Framework for Predictive Talent Management

Chintakunta Sai Meghana
MBA(General) Research Scholar, Regd No. 240A1E0036

JNTUA School of Management Studies, Jawaharlal Nehru Technological University, Anantapuramu - 515001, AP, India

Under the Guidance of
Dr P Sravan Kumar

,Asst Professor(A)

JNTUA SCHOOL OF MANAGEMENT STUDIES,

JNT University ANANTAPUR dr.psravan@gmail.com

Correspondence: meghanaanu2002@gmail.com | 2024-2026

Abstract: Purpose: India's workforce, numbering over 500 million, faces unprecedented transformation pressures as Industry 4.0 technologies reshape the employment landscape. Yet the country lacks an open-source, culturally contextualized predictive analytics framework that organizations can readily adopt for talent management decision-making. This paper addresses that gap by developing and validating an end-to-end analytics pipeline tailored to Indian workforce dynamics.

Methodology: We employ a quantitative, cross-sectional research design using a synthetically generated dataset of 1,200 employee records spanning 30 variables across six predictor categories: demographics, job characteristics, engagement metrics, well-being indicators, career development factors, and compensation structures. Four supervised machine learning models—Logistic Regression, Random Forest, Gradient Boosting, and XGBoost—are trained and evaluated using stratified 5-fold cross-validation. The analytical framework encompasses data preprocessing, feature engineering with seven derived variables, and comprehensive model benchmarking.