Abstract:

Introduction: Rwanda reported the rate at 33% in 2020. Globally, deaths from malnutrition

stand at 45% of child deaths. According to the World Health Organization's (WHO) goal to

reduce malnutrition by 3.9% per year, all countries must define appropriate strategies. Data

science and machine learning are currently important contributors to predicting and assessing the

outcomes of patients. Hence, developing a model capable of predicting stunting in Rwandan

children is critical, so that the best options for early diagnosis or detection of stunting, as well as

treatment, can be made a policy priority for the community and health services.

Methods: The Rwanda Demographic and Health Survey (RDHS) 2019-2020 was utilized as

secondary data. Using stratified 10-fold cross-validation, different Machine learning classifiers

were trained to predict stunting status. The prediction models were compared using a variety of

metrics, including accuracy, sensitivity, specificity, F1 score, and AUC, and the best model was

chosen.

Results: The Gradient Boosting Classifier developed the best model, revealing that the training

accuracy was 80.49 % based on the performance indicators of several models. Based on the

confusion matrix, the test accuracy, sensitivity, specificity, and F1 were calculated, yielding the

model's ability to classify stunting cases correctly at 79.33%, identify stunted children correctly

at 72.51%, and categorize non-stunted children correctly at 94.49%, with an area under the curve

of 0.89. The model found that the mother's height, television watching, the child's age, province,

mother education, birth weight childbirth size were the most important predictors of stunting

status.

Recommendation: Therefore, machine-learning techniques may be used not only in Rwanda but

also worldwide to construct an accurate model that can detect the early stages of stunting and

offer the best predictive attributes to help in the prevention and control of stunting in Rwandan

children under the age of five.

Abstract:

Abstract:

Abstract:

Abstract:

Abstract: