Predicting Global Under-five Mortality Trends Using Machine Learning Based on the World Health Organization Data: A Cross-country Analysis
Nurzhanat Ussengazy
A
BSTRACT
Objective:
To analyze global trends in under-five mortality and develop a predictive model using machine learning based on the World Health Organization (WHO) Global Health Observatory data.
Materials and Methods:
A cross-country dataset, including under-five mortality rates, country, year, and sex, was obtained from the WHO. Data were cleaned and analyzed using the descriptive statistics. A random forest regression model was applied to predict the mortality rates. Model performance was evaluated using mean absolute error (MAE) and the coefficient of determination (
Results:
A substantial global decline in under-five mortality was observed over time. The random forest model demonstrated high predictive accuracy (MAE = 5.82;
Conclusion:
Under-five mortality has declined globally but remains unevenly distributed. Machine learning models provide accurate predictions and highlight key determinants, supporting their use in public health planning.