A Neural Network–Based Approach to Analyze the Impact of Talent Management on Employee Retention in Talent Acquisition

Abstract

This article presents a neural network (NN)–based study that quantifies how talent management practices affect employee retention during talent acquisition phases. Using a dataset of 400 newly hired employees across three industries, we build and evaluate a multilayer perceptron that predicts retention (stayed ≥ 12 months) from talent-management-related features. The NN outperforms standard baselines (logistic regression and random forest), achieving an accuracy of 85% and AUC = 0.91, indicating strong predictive power and evidence that specific talent-management interventions are associated with higher retention.