Measuring non-monetary poverty via machine learning and neutrosophic method: Review

Abstract

Listen

Translate article

Poverty is an emerging problem that most economies are facing today. The study is aimed at exploring research conducted on measuring non-monetary poverty via machine learning. Non-monetary poverty is identified through the following factors: demographics, population, distribution of income, climate, culture, ethnics, and availability of natural and artificial resources. Today, one of the most important aspects of non-monetary poverty measurement is using machine learning for multiple data points other than wealth or income to assess the quality of life of an individual or community. The socioeconomic factors that contribute poverty in emerging nations have also been found using machine learning algorithms. To achieve our goal neutrosophic model and machine learning algorithms were applied. Neutrosophic model used for reviewing the poverty indicators along with ML algorithms. While exploring the utility of machine learning in our study to measure poverty we will find the answers for the following questions: (1) Why it is important to take into consideration of non-monetary approaches while calculating poverty rate? (2) Which machine learning algorithms were used in poverty measurement? (3) What is the future scope of machine learning applications in poverty prediction? In finding answers for those questions, we have analyzed overall 10 papers which were collected according to exclusion and inclusion criteria and the purpose of the selection according to the content of the paper. During the survey it was found out that machine learning gives sophisticated data for identifying non-monetary reasons of poverty and this survey is first that uses machine learning to non-monetary poverty factors.