Detecting banking frauds with analytics and machine learning
Bank fraud is the bodily loss of a Bank or maybe the loss of very sensitive info. For detection, there are lots of machine learning algorithms which can be used. The study shows many algorithms which could be used for deciding transactions as fraud or perhaps real. The information set employed in Ba...
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
Czech Technical University in Prague
2023-06-01
|
Series: | Business & IT |
Subjects: | |
Online Access: | http://bit.fsv.cvut.cz/issues/01-23/full_01-23_11.pdf |
_version_ | 1797400566419161088 |
---|---|
author | Daniella Maya Haddab |
author_facet | Daniella Maya Haddab |
author_sort | Daniella Maya Haddab |
collection | DOAJ |
description | Bank fraud is the bodily loss of a Bank or maybe the loss of very sensitive info. For detection, there are lots of machine learning algorithms which can be used. The study shows many algorithms which could be used for deciding transactions as fraud or perhaps real. The information set employed in Bank fraud Detection was utilized in the research. The SMOTE method was used for oversampling, since the dataset was incredibly imbalanced. Moreover, include choice was performed, and the set was divided into two parts, test data and instruction information. The algorithms used in this study were Logistic Regression, Multilayer Perceptron, Random Forest and Naive Bayes. The results show that every algorithm could be used with good precision for fraud detection of banking solutions. For the detection of extra constipation, the proposed model might be used. |
first_indexed | 2024-03-09T01:57:25Z |
format | Article |
id | doaj.art-1333303d49314392b9416fa619d2d042 |
institution | Directory Open Access Journal |
issn | 1805-3777 2570-7434 |
language | English |
last_indexed | 2024-03-09T01:57:25Z |
publishDate | 2023-06-01 |
publisher | Czech Technical University in Prague |
record_format | Article |
series | Business & IT |
spelling | doaj.art-1333303d49314392b9416fa619d2d0422023-12-08T12:49:28ZengCzech Technical University in PragueBusiness & IT1805-37772570-74342023-06-01131909610.14311/bit.2023.01.11Detecting banking frauds with analytics and machine learningDaniella Maya Haddab0Weizman Institue of ScienceBank fraud is the bodily loss of a Bank or maybe the loss of very sensitive info. For detection, there are lots of machine learning algorithms which can be used. The study shows many algorithms which could be used for deciding transactions as fraud or perhaps real. The information set employed in Bank fraud Detection was utilized in the research. The SMOTE method was used for oversampling, since the dataset was incredibly imbalanced. Moreover, include choice was performed, and the set was divided into two parts, test data and instruction information. The algorithms used in this study were Logistic Regression, Multilayer Perceptron, Random Forest and Naive Bayes. The results show that every algorithm could be used with good precision for fraud detection of banking solutions. For the detection of extra constipation, the proposed model might be used.http://bit.fsv.cvut.cz/issues/01-23/full_01-23_11.pdfbanking fraudlogistic regressionrandom forest |
spellingShingle | Daniella Maya Haddab Detecting banking frauds with analytics and machine learning Business & IT banking fraud logistic regression random forest |
title | Detecting banking frauds with analytics and machine learning |
title_full | Detecting banking frauds with analytics and machine learning |
title_fullStr | Detecting banking frauds with analytics and machine learning |
title_full_unstemmed | Detecting banking frauds with analytics and machine learning |
title_short | Detecting banking frauds with analytics and machine learning |
title_sort | detecting banking frauds with analytics and machine learning |
topic | banking fraud logistic regression random forest |
url | http://bit.fsv.cvut.cz/issues/01-23/full_01-23_11.pdf |
work_keys_str_mv | AT daniellamayahaddab detectingbankingfraudswithanalyticsandmachinelearning |