Credit Card Fraud Detection

Fuente: Zenodo
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: C.Aarthi, A.Ajisha, Dr.R.Ravi
Format: Recurso digital
Veröffentlicht: Zenodo 2025
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866902042727940096
author C.Aarthi, A.Ajisha, Dr.R.Ravi
author_facet C.Aarthi, A.Ajisha, Dr.R.Ravi
contents <p> The banking industry is seriously threatened by credit card fraud, as illegal transactions result in<br>billions of dollars being lost annually worldwide.This study presents a machine learning-based method for<br>detecting credit card fraud in real-time. The suggested method makes use of an interactive web application<br>created with Streamlit that incorporates a pre-trained classification model. Thirty numerical features that<br>reflect transaction characteristics are used to train the model using anonymised transaction data. Through<br>the interface, users enter transaction details, and the system classifies the transaction as either legitimate or<br>fraudulent based on its prediction of the likelihood of fraud. By spotting obscure patterns in the feature<br>space, the model shows excellent accuracy in finding anomalies. The purpose of this application is to help<br>analysts and financial institutions quickly determine the legality of transactions and reduce the risks<br>associated with fraud. The system is a useful tool for both industry and research because it is portable, easy<br>to use, and deployable across multiple platforms. </p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_15400188
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Credit Card Fraud Detection
C.Aarthi, A.Ajisha, Dr.R.Ravi
<p> The banking industry is seriously threatened by credit card fraud, as illegal transactions result in<br>billions of dollars being lost annually worldwide.This study presents a machine learning-based method for<br>detecting credit card fraud in real-time. The suggested method makes use of an interactive web application<br>created with Streamlit that incorporates a pre-trained classification model. Thirty numerical features that<br>reflect transaction characteristics are used to train the model using anonymised transaction data. Through<br>the interface, users enter transaction details, and the system classifies the transaction as either legitimate or<br>fraudulent based on its prediction of the likelihood of fraud. By spotting obscure patterns in the feature<br>space, the model shows excellent accuracy in finding anomalies. The purpose of this application is to help<br>analysts and financial institutions quickly determine the legality of transactions and reduce the risks<br>associated with fraud. The system is a useful tool for both industry and research because it is portable, easy<br>to use, and deployable across multiple platforms. </p>
title Credit Card Fraud Detection
url https://doi.org/10.5281/zenodo.15400188