Thesis CLASIFICACIÓN DE SOLICITANTES COMO APOYO AL ANÁLISIS DE RIESGO EN CUMPLO CHILE S.A.
Loading...
Date
2018
Authors
Journal Title
Journal ISSN
Volume Title
Program
Campus
Abstract
Hoy en día las instituciones bancarias están usando credit scores para evaluar el potencial riesgo potencial que representa prestar dinero a los clientes, y mitigar las pérdidas debidas a deudas incobrables. Un credit score es una expresión numérica que representa la solvencia un individuo, es decir, su capacidad de satisfacer las deudas que adquiere. Estos puntajes son obtenidos principalmente mediante la interpretación y análisis de los grandes volúmenes de datos que poseen los organismos financieros. Los prestadores utilizan los credit scores para determinar quién califica para un préstamo, a qué tasa de interés y a qué plazo, así como también para identificar aquellos clientes que serán más rentables.En este documento se presenta un estudio de minería de datos realizado para la empresa de crowdfunding financiero Cumplo Chile S.A., el cual tiene como objetivo apoyar el análisis de riesgo que se realiza dentro de la organización. Este estudio fue llevado a cabo aplicando la metodología de trabajo CRISP-DM, y para cumplir con los objetivos del trabajo se utilizaron los algoritmos de clasificación redes neuronales artificiales y convolucionales.
Nowadays, banking institutions are using credit scores to assess the potential risk of lending money to clients, and mitigate losses due to bad debts. A credit score is a numerical expression that represents the solvency of an individual, that is, his ability to satisfy the debts he acquires. These scores are obtained mainly through the interpretation and analysis of the large volumes of data held by financial organizations. This paper presents a data mining study carried out for the financial crowdfunding company Cumplo Chile S.A., which aims to support the risk analysis that is carried out within the organization. This study was carried out applying the framework CRISP-DM, and to meet the objectives of the work, the classification algorithms artificial and convolutional neural networks were used.
Nowadays, banking institutions are using credit scores to assess the potential risk of lending money to clients, and mitigate losses due to bad debts. A credit score is a numerical expression that represents the solvency of an individual, that is, his ability to satisfy the debts he acquires. These scores are obtained mainly through the interpretation and analysis of the large volumes of data held by financial organizations. This paper presents a data mining study carried out for the financial crowdfunding company Cumplo Chile S.A., which aims to support the risk analysis that is carried out within the organization. This study was carried out applying the framework CRISP-DM, and to meet the objectives of the work, the classification algorithms artificial and convolutional neural networks were used.
Description
Catalogado desde la version PDF de la tesis.
Keywords
CREDIT SCORE, CRISP-DM, INDICES COMPRIMIDOS, INSTITUCIONES FINANCIERAS, MINERIA DE DATOS