Skip to content
International Academy of Science, Technology, Engineering and Management (IASTEM)

Conference Proceedings

IASTEM INTERNATIONAL CONFERENCE, Barcelona, Spain 11-11-2025

← All proceedings

Proceedings cover: IASTEM INTERNATIONAL CONFERENCE, Barcelona, Spain 11-11-2025

IASTEM INTERNATIONAL CONFERENCE

Barcelona, Spain 11-11-2025

1 paper

  1. Data-Driven Prediction of Maternal Nutritional Status in Ethiopia Using Ensemble Machine Learning Models

    Amsalu Tessema Adgeh, Simachew Alamneh Aragaw, Tizazu Bayih Beyene, Kassahun Azezew Yidagn, Ayenew Kassie Adimas

    Paper ID IA-ICSTMBARC-111125-17085 · Pages 9-18

    Abstract

    Malnutrition among pregnant women poses a critical challenge to public health in Ethiopia, affecting maternal well-being and increasing the risk of adverse pregnancy outcomes. Despite significant efforts, traditional statistical methods used in nutritional assessments often fall short in capturing complex, multidimensional factors that influence nutritional status. The study aims to develop a predictive model using ensemble machine learning techniques to assess the nutritional status of pregnant women in Ethiopia, leveraging data from the Ethiopian demographic and health survey, managed by the central statistics agency. The research uses a dataset comprising 18,108 records spanning five survey years (2005–2020), with 30 attributes reflecting key socio-demographic and health indicators. A rigorous data preprocessing phase was undertaken, including handling missing values, normalization, and data balancing using the SMOTE technique. Feature selection was conducted using ensemble methods to identify the most influential predictors. The study employs multiple supervised ensemble machine learning algorithms, including XGBoost, Random Forest, CatBoost, and AdaBoost, to develop and compare classification models. Among the tested models, Random Forest model delivered the highest performance and robustness, successfully classifying nutritional status into four categories normal, moderate malnutrition, severe malnutrition, and overnutrition with an accuracy of 97.87%, precision of 97.88%, recall of 97.87%, F1-score of 97.87%, and an outstanding ROC AUC of 99.86%. The results underscore the feasibility of ensemble machine learning in capturing nuanced patterns from complex datasets, offering timely insights for targeted interventions. The findings offer valuable implications for healthcare professionals, policymakers, and researchers by enabling early detection of nutritional risks and informing data-driven strategies to improve maternal health outcomes across Ethiopia. Keywords - Nutritional Status, Pregnant Women, Ensemble Machine Learning, Ethiopian Demographic and Health Survey EDHS, Random Forest, Ethiopia

    Download PDF: Data-Driven Prediction of Maternal Nutritional Status in Ethiopia Using Ensemble Machine Learning Models