Proposed application of Multidimensional Poverty Index (MPI) in Microfinance Industries using Machine Learning

As Microfinance operations are focused in financially under-penetrated low-income regions and have responsibilities of impact investments, they focus on the financial inclusion of the underserved. As the customers served by these organizations are part of the most vulnerable section of society, it is imperative to understand the different aspects of poverty that affect the financial health of a potential or existing debtor. United nations have recognized that eradicating poverty and other deprivations must go hand-in-hand with global strategies to improve health and education, to diminish inequality and boost economic growth. Till now, since the analysis of the creditworthiness of the customer or a group was mostly dependent on the income data and credit history, loan products were also designed depending only on these aspects. Taking account of the different social or economic aspects while screening a new business area or designing their loan products has become crucial for the Microfinance sector. The objective of this research is to predict and rate the depth of multidimensional poverty of the residents of a geographical area depending upon the deprivation faced by residents and to build a product recommendation system which can help in alleviating the deprivations faced by the customer. The dataset used was obtained from the MPI website for 2018.