Machine Learning (ML) Technologies for Digital Credit Scoring in Rural Finance: A Literature Review

Abstract

Rural credit is one of the most critical inputs for farm production across the globe. Despite so many advances in digitalization in emerging and developing economies, still a large part of society like small farm holders, rural youth, and women farmers are untouched by the mainstream of banking transactions. Machine learning-based technology is giving a new hope to these individuals. However, it is the banking or non-banking institutions that decide how they will adopt this advanced technology, to have reduced human biases in loan decision making. Therefore, the scope of this study is to highlight the various AI-ML- based methods for credit scoring and their gaps currently in practice by banking or non-banking institutions. For this study, systematic literature review methods have been applied; existing research articles have been empirically reviewed with an attempt to identify and compare the best fit AI-ML-based model adopted by various financial institutions worldwide. The main purpose of this study is to present the various ML algorithms highlighted by earlier researchers that could be fit for a credit assessment of rural borrowers, particularly those who have no or inadequate loan history. However, it would be interesting to recognize further how the financial institutions could be able to blend the traditional and digital methods successfully without any ethical challenges.

Analysis of Factors Influencing Credit Access of Vietnamese Informal Labors in the Time of COVID-19 Pandemic

“…Besides these traditional models, some recent researches have used machine learning to utilize all data available and systematic models that are accurate, efficient, and objective. In term of credit scoring, Kumar et al (2021) use systematic literature review methods to identify and compare the best fit AI-ML-based model adopted by various financial institutions worldwide. The main purpose of this study is to present the various ML algorithms highlighted by earlier researchers that could be fit for a credit assessment of rural borrowers, particularly those who have no or inadequate loan history.…”

Impacts of Digital Labor on Allocation of Economic Resources in a Society on the Background of Artificial Intelligence

“…In some circumstances, artificial intelligence and digital labor can be used to support development of rural areas. For example, Kumar et al (2021) found that machine learning technologies could facilitate digital credit scoring in rural finance so that people living in rural areas were more likely to access to financial services. In fact, people living in rural areas usually have limited access to financial services and goods.…”

Financial Inclusion of Vulnerable Sectors with a Gender Perspective: Risk Analysis Model with Artificial Intelligence Based on Complex Thinking

“…We propose a gendersensitive risk analysis model, speci cally targeting impoverished and vulnerable women, using AI within the framework of complex thinking to achieve the nancial inclusion of an under-nanced population segment through formal nancial institutions. Our proposal differs from others because the specialized literature on the subject (Kumar et al, 2021b;Akter et al, 2021) points to the need for new variables that integrate gender, poverty and vulnerability, and entrepreneurship to create speci c products and services but do not contemplate complex thinking; although some models use AI and consider the clients' contextual complexity, they do not include the framework of complex thinking.…”

“…Low-income people do not necessarily have higher credit risks than the banked population because they can transform the three basic types of available resources (physical capital, human capital, and social capital) into assets to improve their quality of life. Currently, AI and machine learning-based technologies have designed new ways to do credit scoring from customer characteristics (the complete pro le of the borrower's current income level, employment opportunities, and potential ability to repay the loan) (Kumar et al, 2021;Kumar et al, 2021b); also, there are non-traditional combinations of various models and algorithms to design better options for credit scoring (Papouskova and Hajek, 2019;Xu et al, 2020;Liu et al, 2022). New models have improved the statistical performance of credit rating models, some incorporating alternative data sources such as calls made from cell phones, mobile device ngerprints, e-commerce, social networks, email, psychometric variables, and the integration of public and mobile geospatial data from satellites; that is, non- This article proposes a comprehensive solution to promote access and permanence in the entrepreneurial ecosystem for women who run micro businesses and are in a situation of poverty and vulnerability in Mexico.…”