![](data:image/png;base64,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)
Jones-Ortiz, Guzmán-Seraquive.
Análisis de las técnicas de machine learning aplicadas en la detección de fraudes bancarios
from no more than two years ago. It seeks to eliminate the need for expert
knowledge in data analysis, however, human intuition remains a key element
and cannot be replaced in its entirety. The objective of the article is to analyze
the most used machine learning techniques in the detection of bank fraud,
knowing the metrics used.
Introducción
Con el pasar de los años, la inteligencia artificial ha evolucionado al punto de
presentar diferentes metodologías de aprendizaje automático aplicadas a un
sinnúmero de áreas en la vida cotidiana; actualmente, con la gran cantidad de
documentos que se producen y publican en la web, es fundamental contar con
herramientas tecnológicas que permita a las personas obtener, procesar y
discernir información que le resulte de utilidad en su formación profesional. “Las
técnicas de aprendizaje automático están experimentando un auge sin
precedentes en diversos ámbitos, tanto en el mundo académico como en el
empresarial, constituyendo una palanca de transformación relevante” (Calvo,
Guzmán, & Ramos, 2018, p. 5). Con el desarrollo tecnológico que se evidencia,
en el área de educación, empresarial, y con mucha presencia en los últimos
años en temas de seguridad, el aprendizaje automático o de máquina, ha
tomado principal relevancia con el objetivo de alcanzar un mayor y mejor
entendimiento de toda la información que reposa en la nube.
Al ser esta una tecnología basada en patrones, es capaz de aprender relaciones
y tendencias de manera automática, permitiendo integrar técnicas de gran
capacidad de analítica como el machine learning, es posible entre otras cosas
monitorear y configurar parámetros que colaboren en la detección de acciones
que antes eran más difícil de prevenir. En la actualidad, diferentes empresas,
sobre todo del área financiera, optan por esta alternativa para evaluar
escenarios donde se evalúen clientes con perfiles riesgosos e inclusive
determinar operaciones que signifiquen fraudes.
El objetivo del artículo es analizar las técnicas de machine learning mas
utilizadas para la detección de fraude bancaria mediante una revisión de
literatura generando conocimiento relacionado a las métricas empleadas.
Materiales y métodos
Se empleó una revisión de literatura con el fin de evidenciar las técnicas
empleadas para la detección de fraudes bancarios, se revisaron artículos desde
diferentes fuentes bibliográficas publicado en los años 2018 al 2020 con la
finalidad de recabar información lo más actualizada posible. El objetivo es
analizar la popularidad, características y eficiencia de las principales técnicas
de minería de datos y machine learning desde el punto de vista que los
investigadores exponen en función de la detección de fraudes bancarios.
Para garantizar la calidad y veracidad de la información consultada se aplican
estrategias de búsqueda que garanticen un adecuado proceso de selección, se
utilizaron bases de datos como Scopus, Scielo, Web of Science o IEEE Xplore;
la selección se limita en función del título, palabras clave y el resumen que
presenta cada artículo. Se evalúa el nivel de detalle que cada artículo ofrece,
115
Revista Científica Ciencia y Tecnología Vol 22 No 33 págs. 114-122
http://cienciaytecnologia.uteg.edu.ec
para ello se plantean preguntas que permiten identificar escenarios positivos y
relevantes para la investigación como:
¿El artículo está enfocado en la detección de fraudes en entidades bancarias?
¿En el artículo se analizan y desarrollan una o más técnicas de machine
learning?
¿Los autores utilizan métricas para la evaluación de la técnica utilizada?
En su mayoría las propuestas analizadas cumplen con las necesidades
planteadas al detallar los procesos de detección de fraudes y la manera en que
es posible evaluar su rendimiento.
Resultados y discusiones
Machine Learning en el sector bancario
El Machine Learning es una metodología para el análisis de datos que facilita el
desarrollo de modelos analíticos. Hueso (2019) indica que “se refiere a la
capacidad que tienen los ordenadores de aprender a partir de los datos,
mediante el uso de algoritmos que permiten a la máquina cambiar su
comportamiento” (pág. 20). Es una rama de la inteligencia artificial
fundamentada en la idea que un sistema puede aprender de datos para así
identificar patrones y tomar decisiones con una mínima intervención del agente
humano.
Siddhant, Anish, Namita, & Arvind (2020) mencionan que esta disciplina se
encuentra estrechamente ligada a la inteligencia artificial, la misma se plantea
con la finalidad de abordar sistemas capaces de aprender automáticamente en
un contexto de identificar patrones y conductas en una amplia base de datos.
El surgimiento del interés en el aprendizaje automático y el constante desarrollo
de las tecnologías de la información y computación ha permitido que esta
técnica evolucione a un contexto iterativo, ya que a medida que los modelos
son expuestos a nuevos datos, éstos pueden aprender y adaptarse de forma
independiente a través de cálculos previos, emitiendo resultados confiables.
Las entidades bancarias persiguen como meta primordial la maximización de
utilidades y reducción de costos; para ello se enfocan en solventar factores
ligados a aspectos sociales, políticos, económicos y tecnológicos (Fernández,
2019). En este último punto es común encontrar que, si bien los recursos
tecnológicos significan un aporte en el desarrollo y mejora en la gestión de
bienes y servicios hacia el cliente, son susceptibles también a generar formas
de cometer delitos, principalmente lo que se conoce como fraude.
Dichos actos ilícitos desembarcan en pérdidas para los clientes y las entidades
bancarias que además pierde credibilidad afectando a la larga su reputación y
posicionamiento como empresa (Giraldo & Caimàn, 2019). El Machine Learning
ha permitido el desarrollo e integración constante de alternativas que permite
monitorear y controlar dichos ataques como por ejemplo sistemas de
verificación de dirección, verificación de tarjeta.
Las herramientas con las que cuentan la inteligencia artificial hacen posible que,
en el sector empresarial de forma específica en entidades financieras, se pueda
elaborar una revisión de gran volumen de datos y de forma rápida (Henderson
& Richard, 2020).
116