En 5 día(s), 5 hora(s) y 11 minuto(s): El Repositorio Institucional UNAL informa a la comunidad universitaria que, con motivo del periodo de vacaciones colectivas, el servicio de publicación estará suspendido: Periodo de cierre: Del 20 de diciembre al 18 de enero de 2026. Sobre los depósitos: Durante este tiempo, los usuarios podrán continuar realizando el depósito respectivo de sus trabajos en la plataforma. Reanudación: Una vez reiniciadas las actividades administrativas, los documentos serán revisados y publicados en orden de llegada.

Pronóstico del cumplimiento de pago de los clientes usando aprendizaje automático

dc.contributor.advisorVelásquez Henao, Juan Davidspa
dc.contributor.authorCampos Cortesía, Zuleyca Carolinaspa
dc.contributor.corporatenameUniversidad Nacional de Colombia - Sede Medellínspa
dc.date.accessioned2020-08-27T21:32:06Zspa
dc.date.available2020-08-27T21:32:06Zspa
dc.date.issued2020-07-11spa
dc.description.abstractThe collection score is presented as a support for the work carried out by the collection area of Créditos Orbe company. The collection score seeks to direct collection resources and strategies to groups of clients determined by their probability of payment, which positively impacts the effectiveness of collection and the efficiency of the process. The literature review shows that only 1.5% of the score models in the last 10 years are focused on collection; These models are generally used for credit placement. In this work, a collection score model is carried out, which is trained based on the historical information of the payment behavior of clients with active credits in the company. The company currently has a logistic regression model, which they use to direct collection strategies. Alternative models are proposed that can improve the precision of the current model, in light of the findings of the literature review. Finally, after training K-nearest neighbor models, logistic regression; and assembly models such as random forests, majority vote, adaboost and bagged with decision trees; the prototype based on random forests performs best, exceeding the metrics of the current model. The prototype is delivered to the company for weekly use, as well as the necessary tools to monitor it and generate performance alertsspa
dc.description.abstractEl score de cartera se presenta como un apoyo para la labor realizada por el área de cartera de la empresa Créditos Orbe. Con el score de cartera, se busca dirigir los recursos y las estrategias de cobro a grupos de clientes determinados por su probabilidad de pago, lo cual impacta positivamente en la eficacia del recaudo y la eficiencia del proceso. La revisión de literatura expone que solo el 1,5% de los modelos de score en los últimos 10 años están enfocados en la cobranza; estos modelos son usados generalmente para la colocación del crédito. En este trabajo se realiza un modelo para score de cartera el cual es entrenado a partir de la información histórica de comportamiento de pago de los clientes con créditos activos en la empresa. La empresa actualmente cuenta con un modelo de regresión logística, el cual usan para dirigir estrategias de cobranza. Se plantean modelos alternativos que pueden mejorar la precisión del modelo actual, a la luz de los hallazgos de la revisión de literatura. Finalmente, después de entrenar modelos de vecino más cercano, regresión logística; y modelos de ensamble como bosques aleatorios, voto mayoritario, adaboost y bagged con árboles de decisión; el prototipo basado en bosques aleatorios es el que mejor se desempeña, superando las métricas del modelo actual. Se le hace entrega a la empresa del prototipo para su uso semanal, así como también, de las herramientas necesarias para monitorearlo y generar alertas de rendimientospa
dc.description.degreelevelMaestríaspa
dc.format.extent81spa
dc.format.mimetypeapplication/pdfspa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/78297
dc.language.isospaspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Medellínspa
dc.publisher.programMedellín - Minas - Maestría en Ingeniería - Analíticaspa
dc.relation.referencesAbe, N., Kowalczyk, M., Domick, M., Gardinier, T., Melville, P., Pendus, C., … Cooley, B. R. (2010). Optimizing debt collections using constrained reinforcement learning. Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD ’10, 75. https://doi.org/10.1145/1835804.1835817spa
dc.relation.referencesAnagnostopoulos, T. T., & Skourlas, C. (2014). Ensemble Majority Voting Classifier for Speech Emotion Recognition and Prediction. (August). https://doi.org/10.1108/JSIT-01-2014-0009spa
dc.relation.referencesAttigeri, G., M, M. P. M., & Pai, R. M. (2019). Framework to predict NPA/Willful defaults in corporate loans: a big data approach. International Journal of Electrical and Computer Engineering (IJECE), 9(5), 3786–3797. https://doi.org/10.11591/IJECE.V9I5.PP%Pspa
dc.relation.referencesBastos, J. A. (2014). Ensemble Predictions of Recovery Rates. Journal of Financial Services Research, 46(2), 177–193. https://doi.org/10.1007/s10693-013-0165-3spa
dc.relation.referencesBellotti, T., & Crook, J. (2009). Credit scoring with macroeconomic variables using survival analysis. Journal of the Operational Research Society, 60(12), 1699–1707. https://doi.org/10.1057/jors.2008.130spa
dc.relation.referencesBoz, Z., Gunnec, D., Birbil, S. I., & Öztürk, M. K. (2018). Reassessment and Monitoring of Loan Applications with Machine Learning. Applied Artificial Intelligence, 32(9–10), 939–955. https://doi.org/10.1080/08839514.2018.1525517spa
dc.relation.referencesCao, G., Downes, A., Khan, S., Wong, W., & Xu, G. (2018). Taxpayer Behavior Prediction in SMS Campaigns. 2018 5th International Conference on Behavioral, Economic, and Socio-Cultural Computing (BESC), 19–23. https://doi.org/10.1109/BESC.2018.8697317spa
dc.relation.referencesChavarín Rodríguez, R. (2015). Revista mexicana de economía y finanzas. In Revista mexicana de economía y finanzas (Vol. 10). Retrieved from http://www.scielo.org.mx/scielo.php?script=sci_arttext&pid=S1665-53462015000100073spa
dc.relation.referencesCheng, D., & Cirillo, P. (2018). A reinforced urn process modeling of recovery rates and recovery times. Journal of Banking & Finance, 96, 1–17. https://doi.org/10.1016/j.jbankfin.2018.08.014spa
dc.relation.referencesCunningham, P., & Delany, S. J. (2014). k-Nearest neighbour classifiers k -Nearest Neighbour Classifiers. (April 2007).spa
dc.relation.referencesDirick, L., Bellotti, T., Claeskens, G., & Baesens, B. (2019). Macro-Economic Factors in Credit Risk Calculations: Including Time-Varying Covariates in Mixture Cure Models. Journal of Business & Economic Statistics, 37(1), 40–53. https://doi.org/10.1080/07350015.2016.1260471spa
dc.relation.referencesDirick, L., Claeskens, G., & Baesens, B. (2017). Time to default in credit scoring using survival analysis: a benchmark study. Journal of the Operational Research Society, 68(6), 652–665. https://doi.org/10.1057/s41274-016-0128-9spa
dc.relation.referencesDressler, J. B., & Stokes, J. R. (2010). Survival analysis and mortgage termination at AgChoice ACA. Agricultural Finance Review, 70(1), 21–36. https://doi.org/10.1108/00021461011042611spa
dc.relation.referencesGórajski, M., Serwa, D., & Wośko, Z. (2019). Measuring expected time to default under stress conditions for corporate loans. Empirical Economics, 57(1), 31–52. https://doi.org/10.1007/s00181-018-1435-6spa
dc.relation.referencesHector, C. (2011). Debt Collection in the Information Age : New Technologies and the Fair Debt Collection Practices Act Debt Collection in the Information Age : New Technologies and the Fair Debt Collection Practices Act. 99(6).spa
dc.relation.referencesJames, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). Springer Texts in Statistics An Introduction to Statistical Learning.spa
dc.relation.referencesJiang, C., Wang, Z., & Zhao, H. (2019). A prediction-driven mixture cure model and its application in credit scoring. European Journal of Operational Research, 277(1), 20–31. https://doi.org/10.1016/j.ejor.2019.01.072spa
dc.relation.referencesKajdanowicz, T., Plamowski, S., Kazienko, P., & Indyk, W. (2012). Transfer Learning Approach to Debt Portfolio Appraisal. https://doi.org/10.1007/978-3-642-28931-6_5spa
dc.relation.referencesKim, J., & Kang, P. (2016). Late payment prediction models for fair allocation of customer contact lists to call center agents. Decision Support Systems, 85, 84–101. https://doi.org/10.1016/j.dss.2016.03.002spa
dc.relation.referencesLi, Y., Li, Y., & Li, Y. (2019). What factors are influencing credit card customer’s default behavior in China? A study based on survival analysis. Physica A: Statistical Mechanics and Its Applications, 526, 120861. https://doi.org/10.1016/j.physa.2019.04.097spa
dc.relation.referencesMarimo, M., Clive Malwandla, M., & Gerbrand Breed, D. (2017). A CROSS-SECTIONAL SURVIVAL ANALYSIS REGRESSION MODEL WITH APPLICATIONS TO CONSUMER CREDIT RISK. In South African Statist. J (Vol. 51). Retrieved from https://pdfs.semanticscholar.org/4166/bd323a19ad0256e62e34d64c929610fefa12.pdfspa
dc.relation.referencesOlivella, V. (2018). Panorama de las tendencias de crédito en Colombia Segundo trimestre de 2018spa
dc.relation.referencesOnar, S. C., Oztaysi, B., & Kahraman, C. (2018). a Fuzzy Rule Based Inference System for Early Debt Collection. Technological and Economic Development of Economy, 24(5), 1845–1865. https://doi.org/10.3846/20294913.2016.1266409spa
dc.relation.referencesPeng, C. J., & Ingersoll, G. M. (2014). An Introduction to Logistic Regression Analysis and Reporting. (September 2002). https://doi.org/10.1080/00220670209598786spa
dc.relation.referencesPontificia Universidad Javeriana. Departamento de Administración de Empresas., M. L., & Saavedra García, M. J. (2010). Cuadernos de administración. In Cuadernos de Administración (Vol. 23). Retrieved from http://www.scielo.org.co/scielo.php?script=sci_arttext&pid=S0120-35922010000100013spa
dc.relation.referencesSarlija, N., Bensic, M., & Zekic-Susac, M. (2009). Comparison procedure of predicting the time to default in behavioural scoring. Expert Systems with Applications, 36(5), 8778–8788. https://doi.org/10.1016/j.eswa.2008.11.042spa
dc.relation.referencesShetty, A. S., & Manoj,R. (n.d.). Prediction Of Default Credit Card Users Using Data Mining Techniques. Retrieved from https://www.ijitee.org/wp-content/uploads/papers/v8i7/G5754058719.pdfspa
dc.relation.referencesSo, M. C., Mues, C., de Almeida Filho, A. T., & Thomas, L. C. (2019). Debtor level collection operations using Bayesian dynamic programming. Journal of the Operational Research Society, 70(8), 1332–1348. https://doi.org/10.1080/01605682.2018.1506557spa
dc.relation.referencesStachniss, C. (2005). Supervised Learning of Places from Range Data using AdaBoost. (November 2016). https://doi.org/10.1109/ROBOT.2005.1570363spa
dc.relation.referencesSun, Z., Wiering, M. A., & Petkov, N. (2014a). Classification system for mortgage arrear management. IEEE/IAFE Conference on Computational Intelligence for Financial Engineering, Proceedings (CIFEr), (February), 489–496. https://doi.org/10.1109/CIFEr.2014.6924113spa
dc.relation.referencesSun, Z., Wiering, M. A., & Petkov, N. (2014b). Classification system for mortgage arrear management. 2014 IEEE Conference on Computational Intelligence for Financial Engineering & Economics (CIFEr), 489–496. https://doi.org/10.1109/CIFEr.2014.6924113spa
dc.relation.referencesWijewardhana, U. (2018). A Mathematical Model for Predicting Debt Repayment: A Technical Note. Australasian Accounting, Business and Finance Journal, 12(3), 107–115. https://doi.org/10.14453/aabfj.v12i3.8spa
dc.relation.referencesWittlinger, B., Carranza, L., & Mori, T. (2008). Mejores Prácticas en Estrategias de Cobranza. InSight, 26(November), 1–23. https://doi.org/10.1177/2158244017731494spa
dc.relation.referencesYurynets, R., Yurynets, Z., Dosyn, D., & Kis, Y. (n.d.). Risk Assessment Technology of Crediting with the Use of Logistic Regression Model. Retrieved from http://ceur-ws.org/Vol-2362/paper14.pdfspa
dc.rightsDerechos reservados - Universidad Nacional de Colombiaspa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseAtribución-SinDerivadas 4.0 Internacionalspa
dc.rights.spaAcceso abiertospa
dc.rights.urihttp://creativecommons.org/licenses/by-nd/4.0/spa
dc.subject.ddc310 - Colecciones de estadística generalspa
dc.subject.ddc000 - Ciencias de la computación, información y obras generales::003 - Sistemasspa
dc.subject.proposalscore de carteraspa
dc.subject.proposalcollection scoreeng
dc.subject.proposalmachine learningeng
dc.subject.proposalaprendizaje automáticospa
dc.subject.proposalrandom foresteng
dc.subject.proposalbosques aleatoriosspa
dc.subject.proposalgestión de cobranzaspa
dc.subject.proposalcollection managementeng
dc.titlePronóstico del cumplimiento de pago de los clientes usando aprendizaje automáticospa
dc.title.alternativeForecasting customer payment compliance using machine learningspa
dc.typeTrabajo de grado - Maestríaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/masterThesisspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
642171.2020.pdf
Tamaño:
1.19 MB
Formato:
Adobe Portable Document Format
Descripción:
Tesis de Maestría en Ingeniería - Analítica

Bloque de licencias

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
license.txt
Tamaño:
3.8 KB
Formato:
Item-specific license agreed upon to submission
Descripción: