TY - JOUR
T1 - Modelos analíticos para identificar patrones de delitos financieros
T2 - Una revisión sistemática de la literatura
AU - Ortíz, Maritza Murillo
AU - Marín, Lillyana María Giraldo
AU - Villegas, Herman Horacio Jaramillo
AU - Escobar, Carlos César Piedrahita
N1 - Publisher Copyright:
© 2022, Associacao Iberica de Sistemas e Tecnologias de Informacao. All rights reserved.
PY - 2022/4
Y1 - 2022/4
N2 - One of the greatest challenges facing financial institutions today is the risk of financial crimes that are increasingly sophisticated and global in nature, considering the increasing trends of some types of modalities. For this reason, a systematic literature review on the subject was developed to find out which analytical models are the most used and we managed to detect anomalous situations. From this review, it was identified that, thanks to technology and supervised analytical models such as Support Vector Machine (SVM), Neural Networks, among others, many of the threats that exist in the market today can be considerably mitigated and in this way, it is important to prevent million-dollar losses, however, according to the literature, it is important to take into account that one of the main difficulties in detecting fraud or any other financial crime is unbalanced data, since this implies that the results generated probably show a bias towards the majority class and, in extreme cases, may completely ignore the minority class.
AB - One of the greatest challenges facing financial institutions today is the risk of financial crimes that are increasingly sophisticated and global in nature, considering the increasing trends of some types of modalities. For this reason, a systematic literature review on the subject was developed to find out which analytical models are the most used and we managed to detect anomalous situations. From this review, it was identified that, thanks to technology and supervised analytical models such as Support Vector Machine (SVM), Neural Networks, among others, many of the threats that exist in the market today can be considerably mitigated and in this way, it is important to prevent million-dollar losses, however, according to the literature, it is important to take into account that one of the main difficulties in detecting fraud or any other financial crime is unbalanced data, since this implies that the results generated probably show a bias towards the majority class and, in extreme cases, may completely ignore the minority class.
KW - analytical models
KW - Financial crimes
KW - Machine Learning
UR - http://www.scopus.com/inward/record.url?scp=85136270393&partnerID=8YFLogxK
M3 - Artículo
AN - SCOPUS:85136270393
SN - 1646-9895
VL - 2022
SP - 586
EP - 598
JO - RISTI - Revista Iberica de Sistemas e Tecnologias de Informacao
JF - RISTI - Revista Iberica de Sistemas e Tecnologias de Informacao
IS - E49
ER -