Features Explaining Malnutrition in India: A Machine Learning Approach to Demographic and Health Survey Data

India is one of the severely malnourished countries in the world. Under-nutrition is the reason for death among two-third of the 1.04 million deaths among the children under the age of five in the year 2019. Several strategies have been adopted by the Government of India and state governments to minimize the incidents of malnutrition. However, to make the policies effective, it is important to understand the key features explaining malnutrition. Analyzing the Indian Demographic Health Survey Data (IDHS) of the year 2015–2016, this paper attempts to identify causes of four dimensions of malnutrition namely, Height Age Z-score (HAZ), Weight Age Z-score (WAZ), Weight Height Z-score (WHZ) and Body Mass Index (BMI). Using machine learning approach of feature reduction, the paper identifies ten most important features out of available 1341 features in the database for each of the four anthropometric parameters of malnutrition. The features are reduced and ranked using WEKA tool. Results and finding of this research would provide key policy inputs to address malnutrition and related mortality among the children under the age five.