Machine learning approach for multidimensional poverty estimation

In the social sciences, a theoretical analysis has predominated in its research. The scarcity of data and its difficulty in collecting and storing it, has been the main limitation for the social sciences to adopt quantitative approaches. However, the large amount of information generated in recent...

Full description

Saved in:
Bibliographic Details
Main Authors: Mario Esteban Ochoa Guaraca, Ricardo Castro, Alexander Arias Pallaroso, Antonia Machado, Dolores Sucozhañay
Format: Article
Language:English
Published: Escuela Superior Politécnica del Litoral 2021-11-01
Series:Revista Tecnológica
Subjects:
Online Access:https://rte.espol.edu.ec/index.php/tecnologica/article/view/853
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849714369662812160
author Mario Esteban Ochoa Guaraca
Ricardo Castro
Alexander Arias Pallaroso
Antonia Machado
Dolores Sucozhañay
author_facet Mario Esteban Ochoa Guaraca
Ricardo Castro
Alexander Arias Pallaroso
Antonia Machado
Dolores Sucozhañay
author_sort Mario Esteban Ochoa Guaraca
collection DOAJ
description In the social sciences, a theoretical analysis has predominated in its research. The scarcity of data and its difficulty in collecting and storing it, has been the main limitation for the social sciences to adopt quantitative approaches. However, the large amount of information generated in recent years, mainly through the use of the Internet, has allowed the social sciences to include more and more quantitative analysis. This study proposes the use of technologies such as Machine Learning (ML) are the answers to solving this data scarcity. The objective is to estimate the multidimensional poverty index at the personal level in a particular territory of Ecuador by using Machine Learning (ML) regression models based on a limited amount of data for training. Ten ML models are compared, such as linear, regularized, and assembled models and Random Forest performs outstandingly against the other models. An error of 7.5% was obtained in the cross-validation and 7.48% with the test data set. The estimates are compared with statistical approximations of the MPI in a geographical area and it is obtained that the average MPI estimated by the model compared to the average reported by the statistical studies differs by 1%.
format Article
id doaj-art-3f69d30fc49d417891a1d6828e923a6f
institution DOAJ
issn 0257-1749
1390-3659
language English
publishDate 2021-11-01
publisher Escuela Superior Politécnica del Litoral
record_format Article
series Revista Tecnológica
spelling doaj-art-3f69d30fc49d417891a1d6828e923a6f2025-08-20T03:13:44ZengEscuela Superior Politécnica del LitoralRevista Tecnológica0257-17491390-36592021-11-0133210.37815/rte.v33n2.853Machine learning approach for multidimensional poverty estimationMario Esteban Ochoa Guaraca0Ricardo Castro1Alexander Arias Pallaroso2Antonia Machado3Dolores Sucozhañay4Grupo de Investigación en Población y Desarrollo Local Sustentable (PYDLOS), Departamento Interdisciplinario de Espacio y Población (DIEP), Universidad de Cuenca, Cuenca, EcuadorMicrosoftGrupo de Investigación en Población y Desarrollo Local Sustentable (PYDLOS), Departamento Interdisciplinario de Espacio y Población (DIEP) y Facultad de Jurisprudencia y Ciencias Políticas y Sociales, Universidad de Cuenca, Cuenca, Ecuador.Grupo de Investigación en Población y Desarrollo Local Sustentable (PYDLOS), Departamento Interdisciplinario de Espacio y Población (DIEP), Universidad de Cuenca, Cuenca, EcuadorDepartamento Interdisciplinario de Espacio y Población (DIEP), Facultad de Ciencias Económicas y Administrativas, Universidad de Cuenca, Cuenca, Ecuador. In the social sciences, a theoretical analysis has predominated in its research. The scarcity of data and its difficulty in collecting and storing it, has been the main limitation for the social sciences to adopt quantitative approaches. However, the large amount of information generated in recent years, mainly through the use of the Internet, has allowed the social sciences to include more and more quantitative analysis. This study proposes the use of technologies such as Machine Learning (ML) are the answers to solving this data scarcity. The objective is to estimate the multidimensional poverty index at the personal level in a particular territory of Ecuador by using Machine Learning (ML) regression models based on a limited amount of data for training. Ten ML models are compared, such as linear, regularized, and assembled models and Random Forest performs outstandingly against the other models. An error of 7.5% was obtained in the cross-validation and 7.48% with the test data set. The estimates are compared with statistical approximations of the MPI in a geographical area and it is obtained that the average MPI estimated by the model compared to the average reported by the statistical studies differs by 1%. https://rte.espol.edu.ec/index.php/tecnologica/article/view/853random forestsocial sciencesregressionregionlimited dataset
spellingShingle Mario Esteban Ochoa Guaraca
Ricardo Castro
Alexander Arias Pallaroso
Antonia Machado
Dolores Sucozhañay
Machine learning approach for multidimensional poverty estimation
Revista Tecnológica
random forest
social sciences
regression
region
limited dataset
title Machine learning approach for multidimensional poverty estimation
title_full Machine learning approach for multidimensional poverty estimation
title_fullStr Machine learning approach for multidimensional poverty estimation
title_full_unstemmed Machine learning approach for multidimensional poverty estimation
title_short Machine learning approach for multidimensional poverty estimation
title_sort machine learning approach for multidimensional poverty estimation
topic random forest
social sciences
regression
region
limited dataset
url https://rte.espol.edu.ec/index.php/tecnologica/article/view/853
work_keys_str_mv AT marioestebanochoaguaraca machinelearningapproachformultidimensionalpovertyestimation
AT ricardocastro machinelearningapproachformultidimensionalpovertyestimation
AT alexanderariaspallaroso machinelearningapproachformultidimensionalpovertyestimation
AT antoniamachado machinelearningapproachformultidimensionalpovertyestimation
AT doloressucozhanay machinelearningapproachformultidimensionalpovertyestimation