Implementation of a Machine Learning Model to Address Malnutrition in Toddlers

In this accelerated lifestyle, staying healthy is mandatory in everyone’s life. Proper nutrition enriches both physical health and mental health for the people. This encourages the society to live longer and lowers the risks of early death. Children are the key to the future world’s innovation and development. Providing nutritious food to the children will make them strong and smart. These early years pave the path to the cognitive, physical and social well-being life to the children. Many Government policies and schemes aim to provide safety, health and education to children. It is not only for selfless pursuits but to construct a stable and successful future to society. This research study deals with the prediction of the nutrition status of the child through the use of a machine learning model. There are four machine learning models constructed for the prediction processes. The 265 datasets are collected from the Anganwadi in rural areas of Tamil Nadu. The dataset contains anthropometry factors such as age, height, weight and nutritional status. This paper predicts the nutritional status of children aged from 6 months to 59 months. The machine learning models are built based on the classification algorithms: random forest classifier, Gaussian NB classification algorithm, decision tree classification algorithm and support vector classification algorithm. The best-suited model is identified based on precision, recall, f1-score, support and accuracy evaluation metrics. The decision tree classifier model achieved $98 \%$ accuracy in predicting the nutritional status of the child.