HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment

Financial credit risk scoring is of the utmost importance for credit lenders, a well-designed risk management strategy allows lenders to make better decisions in terms of collocation of financial products and credit limits while keeping losses bounded. In this paper, we propose a Heterogeneous Graph...

Full description

Autores:
Acevedo Viloria, Jaime David
Tipo de recurso:
Fecha de publicación:
2021
Institución:
Universidad de los Andes
Repositorio:
Séneca: repositorio Uniandes
Idioma:
spa
OAI Identifier:
oai:repositorio.uniandes.edu.co:1992/55695
Acceso en línea:
http://hdl.handle.net/1992/55695
Palabra clave:
Graph Neural Networks
Credit Risk
Super-App
Alternative Data
Ingeniería
Rights
openAccess
License
https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
id UNIANDES2_16178a108f01cf4ffb2cb648c4a9d451
oai_identifier_str oai:repositorio.uniandes.edu.co:1992/55695
network_acronym_str UNIANDES2
network_name_str Séneca: repositorio Uniandes
repository_id_str
dc.title.eng.fl_str_mv HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
title HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
spellingShingle HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
Graph Neural Networks
Credit Risk
Super-App
Alternative Data
Ingeniería
title_short HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
title_full HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
title_fullStr HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
title_full_unstemmed HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
title_sort HanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environment
dc.creator.fl_str_mv Acevedo Viloria, Jaime David
dc.contributor.advisor.spa.fl_str_mv Correa Bahnsen, Alejandro
dc.contributor.advisor.none.fl_str_mv Valencia Arboleda, Carlos Felipe
dc.contributor.author.spa.fl_str_mv Acevedo Viloria, Jaime David
dc.contributor.jury.spa.fl_str_mv Cabrales Arévalo, Sergio Andrés
Zarruk Valencia, David
dc.subject.keyword.none.fl_str_mv Graph Neural Networks
Credit Risk
Super-App
Alternative Data
topic Graph Neural Networks
Credit Risk
Super-App
Alternative Data
Ingeniería
dc.subject.themes.none.fl_str_mv Ingeniería
description Financial credit risk scoring is of the utmost importance for credit lenders, a well-designed risk management strategy allows lenders to make better decisions in terms of collocation of financial products and credit limits while keeping losses bounded. In this paper, we propose a Heterogeneous Graph Attention Network model that leverages the many interactions found in a Super-App for the Credit Risk assessment of its users with only Alternative Data. Through the proposed model we manage to learn not only for the scarce amount of users with credit history but also from the users with no credit history; promoting financial inclusion in the process. We test our model in a real-life database of 3.5k users drawn from a Super-App, proving that our model achieves an AUC of 0.69 in terms of classifying whether the users will pay back or not.
publishDate 2021
dc.date.issued.none.fl_str_mv 2021
dc.date.accessioned.none.fl_str_mv 2022-02-22T20:10:31Z
dc.date.available.none.fl_str_mv 2022-02-22T20:10:31Z
dc.type.spa.fl_str_mv Trabajo de grado - Maestría
dc.type.driver.spa.fl_str_mv info:eu-repo/semantics/masterThesis
dc.type.version.spa.fl_str_mv info:eu-repo/semantics/acceptedVersion
dc.type.content.spa.fl_str_mv Text
dc.type.redcol.spa.fl_str_mv http://purl.org/redcol/resource_type/TM
status_str acceptedVersion
dc.identifier.uri.none.fl_str_mv http://hdl.handle.net/1992/55695
dc.identifier.pdf.spa.fl_str_mv 25467.pdf
dc.identifier.instname.spa.fl_str_mv instname:Universidad de los Andes
dc.identifier.reponame.spa.fl_str_mv reponame:Repositorio Institucional Séneca
dc.identifier.repourl.spa.fl_str_mv repourl:https://repositorio.uniandes.edu.co/
url http://hdl.handle.net/1992/55695
identifier_str_mv 25467.pdf
instname:Universidad de los Andes
reponame:Repositorio Institucional Séneca
repourl:https://repositorio.uniandes.edu.co/
dc.language.iso.spa.fl_str_mv spa
language spa
dc.rights.uri.*.fl_str_mv https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
dc.rights.accessrights.spa.fl_str_mv info:eu-repo/semantics/openAccess
dc.rights.coar.spa.fl_str_mv http://purl.org/coar/access_right/c_abf2
rights_invalid_str_mv https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.format.extent.spa.fl_str_mv 11 páginas
dc.format.mimetype.spa.fl_str_mv application/pdf
dc.publisher.spa.fl_str_mv Universidad de los Andes
dc.publisher.program.spa.fl_str_mv Maestría en Ingeniería Industrial
dc.publisher.faculty.spa.fl_str_mv Facultad de Ingeniería
dc.publisher.department.spa.fl_str_mv Departamento de Ingeniería Industrial
institution Universidad de los Andes
bitstream.url.fl_str_mv https://repositorio.uniandes.edu.co/bitstreams/2ebf38cd-2ff5-4419-9151-b4ab0010ed80/download
https://repositorio.uniandes.edu.co/bitstreams/835aaff7-1c79-4a10-8bd9-86d2d7a48751/download
https://repositorio.uniandes.edu.co/bitstreams/f1e2bf0e-593d-4d4a-b5d8-d09453a5f550/download
bitstream.checksum.fl_str_mv 3d113dae35eb43c4ba80413ebecdda2d
5a1bd54b1cdfb740a4a5d132830d7733
5a223fe2ccbad01cc0f8ba42a7b0ffb1
bitstream.checksumAlgorithm.fl_str_mv MD5
MD5
MD5
repository.name.fl_str_mv Repositorio institucional Séneca
repository.mail.fl_str_mv adminrepositorio@uniandes.edu.co
_version_ 1812134003908542464
spelling Al consultar y hacer uso de este recurso, está aceptando las condiciones de uso establecidas por los autores.https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdfinfo:eu-repo/semantics/openAccesshttp://purl.org/coar/access_right/c_abf2Correa Bahnsen, Alejandro37c53098-2130-4ce5-b505-46af4c65daf2400Valencia Arboleda, Carlos Felipevirtual::12880-1Acevedo Viloria, Jaime Davidafe76135-c40c-461b-b3b3-cfc0ca5f10ba500Cabrales Arévalo, Sergio AndrésZarruk Valencia, David2022-02-22T20:10:31Z2022-02-22T20:10:31Z2021http://hdl.handle.net/1992/5569525467.pdfinstname:Universidad de los Andesreponame:Repositorio Institucional Sénecarepourl:https://repositorio.uniandes.edu.co/Financial credit risk scoring is of the utmost importance for credit lenders, a well-designed risk management strategy allows lenders to make better decisions in terms of collocation of financial products and credit limits while keeping losses bounded. In this paper, we propose a Heterogeneous Graph Attention Network model that leverages the many interactions found in a Super-App for the Credit Risk assessment of its users with only Alternative Data. Through the proposed model we manage to learn not only for the scarce amount of users with credit history but also from the users with no credit history; promoting financial inclusion in the process. We test our model in a real-life database of 3.5k users drawn from a Super-App, proving that our model achieves an AUC of 0.69 in terms of classifying whether the users will pay back or not.La calificación del riesgo de crédito financiero es de suma importancia para los prestamistas de crédito, una estrategia de gestión de riesgo bien diseñada permite a los prestamistas tomar mejores decisiones en términos de colocación de productos financieros y límites de crédito, manteniendo las pérdidas limitadas. En este artículo, proponemos un modelo de Graph Attention Networks que aprovecha las muchas interacciones encontradas en una Super-App para la evaluación del Riesgo de Crédito de sus usuarios con solo Datos Alternativos. A través del modelo propuesto logramos aprender no solo de la escasa cantidad de usuarios con historial crediticio sino también de los usuarios sin historial crediticio; promover la inclusión financiera en el proceso. Probamos nuestro modelo en una base de datos de la vida real de 3.5k usuarios extraídos de una Super-App, lo que demuestra que nuestro modelo logra un AUC de 0.69 en términos de clasificar si los usuarios pagarán o no.Magíster en Ingeniería IndustrialMaestría11 páginasapplication/pdfspaUniversidad de los AndesMaestría en Ingeniería IndustrialFacultad de IngenieríaDepartamento de Ingeniería IndustrialHanCred: Heterogeneous graph attention networks for credit risk assessment in a Super-App environmentTrabajo de grado - Maestríainfo:eu-repo/semantics/masterThesisinfo:eu-repo/semantics/acceptedVersionTexthttp://purl.org/redcol/resource_type/TMGraph Neural NetworksCredit RiskSuper-AppAlternative DataIngeniería201119605Publicationhttps://scholar.google.es/citations?user=vPH5LywAAAAJvirtual::12880-1https://scienti.minciencias.gov.co/cvlac/visualizador/generarCurriculoCv.do?cod_rh=0000271403virtual::12880-1e1de19e8-629e-401d-a9d3-77eea3d2db48virtual::12880-1e1de19e8-629e-401d-a9d3-77eea3d2db48virtual::12880-1THUMBNAIL25467.pdf.jpg25467.pdf.jpgIM Thumbnailimage/jpeg27476https://repositorio.uniandes.edu.co/bitstreams/2ebf38cd-2ff5-4419-9151-b4ab0010ed80/download3d113dae35eb43c4ba80413ebecdda2dMD53TEXT25467.pdf.txt25467.pdf.txtExtracted texttext/plain60871https://repositorio.uniandes.edu.co/bitstreams/835aaff7-1c79-4a10-8bd9-86d2d7a48751/download5a1bd54b1cdfb740a4a5d132830d7733MD52ORIGINAL25467.pdfapplication/pdf768181https://repositorio.uniandes.edu.co/bitstreams/f1e2bf0e-593d-4d4a-b5d8-d09453a5f550/download5a223fe2ccbad01cc0f8ba42a7b0ffb1MD511992/55695oai:repositorio.uniandes.edu.co:1992/556952024-03-13 14:47:55.16https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdfopen.accesshttps://repositorio.uniandes.edu.coRepositorio institucional Sénecaadminrepositorio@uniandes.edu.co

Publicaciones similares