Leveraging Machine Learning Techniques to Effectively Predict Malnutrition among Children

The objective of the research is to predict malnutrition in children through the application of different algorithms in machine learning. The lack of adequate muscle and fat tissue development during intrauterine growth is a hallmark of malnutrition. It is mostly brought on by inadequate nourishment for the mother and placental insufficiency, which results in lower rates of infant morbidity and mortality across the world. This study focuses on to determine whether malnutrition is present in children by calculating their Z-score, which takes into consideration age in months, weight, peak, and sex. The dataset that was used in this project was obtained for network education from UNICEF. The given dataset is divided into two parts: one for verification and the other for analysis. We identify infant malnutrition by calculating WAZ (underweight) and LAZ (stunting), and we train the fashions to do so. Several machine learning techniques, such as logistic regression, KNN, and Naïve Bayes, are used to identify childhood malnutrition. Out of all these designs, logistic regression shows better accuracy than the other algorithms.