The Application of Machine Learning to the Task of Poverty Classification

For many societal and economic issues, poverty has become a stubborn underlying cause. The high rate of reproduction is a contributing factor to India's poverty. The government must implement a small number of effective programmes if poverty is to be reduced. Most of the time, a survey is conducted prior to the creation of policies, and only direct factors like income or consumption levels are considered. However, direct measures alone are insufficient for classifying households into poverty brackets. Therefore, we divide the poor into categories according to factors such as education, number of people in the home, housing conditions, and many more. (i) we extract a subset of features that are critical for identifying poverty; (ii)We investigate the effects of the extracted features on the category; and (iii)To choose the most reliable machine learning model, we utilize numerous models and evaluate their results. Instead of utilizing the traditional one-dimensional lens to study poverty levels, we employ the "Proxy Means Test" ("PMT") to look at different levels of poverty in a broader context.