Author:
Liu Bocheng,Chen Xiang,Yu Kaizhi
Abstract
Abstract
With the rapid development of Internet technology, the scale of online transactions is constantly expanding. At the same time, the related network transaction fraud problem has become more significant. Compared with the credit card transaction, the network transaction has the characteristics of low cost, wide coverage and high frequency, which makes the detection of fraud more complex. Aiming at the problem of difficult fraud detection in network transactions, this paper designed two fraud detection algorithms based on Fully Connected Neural Network and XGBoost, whose AUC values can achieve 0.912 and 0.969 respectively. Meanwhile, we designed an interactive online transaction fraud detection system based on XGBoost model, which can automatically analyze the transaction data uploaded and return the fraud detection results to users.
Subject
General Physics and Astronomy
Reference10 articles.
1. Graph Computing for Financial Crime and Fraud Detection: Trends, Challenges and Outlook;Kurshan,2020
2. Analysis of credit card fraud detection methods;Dheepa;International journal of recent trends in engineering,2009
3. A model Based on Convolutional Neural Network for Online Transaction Fraud Detection;Zhang;Security and Communication Networks,2018
4. Perspective analysis of telecommunication fraud detection using data stream analytics and neural network classification based data mining;Jain;International Journal of Information Technology,2017
5. Gaussian Error Linear Units (GELUs);Hendrycks,2016
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献
1. Automated Fraud Detection in Online Transaction Using Representation Learning;2024 International Conference on Communication, Computing and Internet of Things (IC3IoT);2024-04-17
2. Risk-based tsunami early warning using random forest;Computers & Geosciences;2023-10