Predictive Analytics for Child Mortality: Leveraging Machine Learning for Early Intervention

Abstract

Child Mortality (CM) remains a critical global issue, particularly in low- and middle-income countries (LMICs). This study employs machine learning models—Logistic Regression, Support Vector Machine (SVM), and Random Forest—to predict child mortality risk using key health indicators. The dataset, preprocessed using feature scaling and encoding techniques, undergoes training and evaluation through multi-class classification. The models are assessed based on accuracy, confusion matrices, ROC curves, and Matthews Correlation Coefficient (MCC) to determine their predictive effectiveness. The proposed approach not only provides insights into child mortality risk factors but also demonstrates the potential of machine learning in healthcare analytics. The developed system can assist healthcare professionals and policymakers in early intervention strategies, ultimately reducing child mortality rates through data-driven decision-making.