A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory
Money laundering is an illicit activity that seeks to conceal the nature and origins of criminal proceeds, posing a substantial threat to the national economy, the political order, and social stability. To scientifically and reasonably predict money laundering risks, this paper focuses on the “layer...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-03-01
|
Series: | Symmetry |
Subjects: | |
Online Access: | https://www.mdpi.com/2073-8994/16/3/378 |
_version_ | 1797239254541139968 |
---|---|
author | Fei Wan Ping Li |
author_facet | Fei Wan Ping Li |
author_sort | Fei Wan |
collection | DOAJ |
description | Money laundering is an illicit activity that seeks to conceal the nature and origins of criminal proceeds, posing a substantial threat to the national economy, the political order, and social stability. To scientifically and reasonably predict money laundering risks, this paper focuses on the “layering” stage of the money laundering process in the field of supervised learning for money laundering fraud prediction. A money laundering and fraud prediction model based on deep learning, referred to as MDGC-LSTM, is proposed. The model combines the use of a dynamic graph convolutional network (MDGC) and a long short-term memory (LSTM) network to efficiently identify illegal money laundering activities within financial transactions. MDGC-LSTM constructs dynamic graph snapshots with symmetrical spatiotemporal structures based on transaction information, representing transaction nodes and currency flows as graph nodes and edges, respectively, and effectively captures the relationships between temporal and spatial structures, thus achieving the dynamic prediction of fraudulent transactions. The experimental results demonstrate that compared with traditional algorithms and other deep learning models, MDGC-LSTM achieves significant advantages in comprehensive spatiotemporal feature modeling. Specifically, based on the Elliptic dataset, MDGC-LSTM improves the Macro-F1 score by 0.25 compared to that of the anti-money laundering fraud prediction model currently considered optimal. |
first_indexed | 2024-04-24T17:48:37Z |
format | Article |
id | doaj.art-a35edd0053674530b26e4ac53e3806b9 |
institution | Directory Open Access Journal |
issn | 2073-8994 |
language | English |
last_indexed | 2024-04-24T17:48:37Z |
publishDate | 2024-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Symmetry |
spelling | doaj.art-a35edd0053674530b26e4ac53e3806b92024-03-27T14:05:38ZengMDPI AGSymmetry2073-89942024-03-0116337810.3390/sym16030378A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term MemoryFei Wan0Ping Li1School of Management, Hefei University of Technology, Hefei 230009, ChinaSchool of Information Engineering, Fuyang Normal University, Fuyang 236037, ChinaMoney laundering is an illicit activity that seeks to conceal the nature and origins of criminal proceeds, posing a substantial threat to the national economy, the political order, and social stability. To scientifically and reasonably predict money laundering risks, this paper focuses on the “layering” stage of the money laundering process in the field of supervised learning for money laundering fraud prediction. A money laundering and fraud prediction model based on deep learning, referred to as MDGC-LSTM, is proposed. The model combines the use of a dynamic graph convolutional network (MDGC) and a long short-term memory (LSTM) network to efficiently identify illegal money laundering activities within financial transactions. MDGC-LSTM constructs dynamic graph snapshots with symmetrical spatiotemporal structures based on transaction information, representing transaction nodes and currency flows as graph nodes and edges, respectively, and effectively captures the relationships between temporal and spatial structures, thus achieving the dynamic prediction of fraudulent transactions. The experimental results demonstrate that compared with traditional algorithms and other deep learning models, MDGC-LSTM achieves significant advantages in comprehensive spatiotemporal feature modeling. Specifically, based on the Elliptic dataset, MDGC-LSTM improves the Macro-F1 score by 0.25 compared to that of the anti-money laundering fraud prediction model currently considered optimal.https://www.mdpi.com/2073-8994/16/3/378money launderingfraud predictiondynamic graph convolutionspatiotemporal feature modeling |
spellingShingle | Fei Wan Ping Li A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory Symmetry money laundering fraud prediction dynamic graph convolution spatiotemporal feature modeling |
title | A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory |
title_full | A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory |
title_fullStr | A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory |
title_full_unstemmed | A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory |
title_short | A Novel Money Laundering Prediction Model Based on a Dynamic Graph Convolutional Neural Network and Long Short-Term Memory |
title_sort | novel money laundering prediction model based on a dynamic graph convolutional neural network and long short term memory |
topic | money laundering fraud prediction dynamic graph convolution spatiotemporal feature modeling |
url | https://www.mdpi.com/2073-8994/16/3/378 |
work_keys_str_mv | AT feiwan anovelmoneylaunderingpredictionmodelbasedonadynamicgraphconvolutionalneuralnetworkandlongshorttermmemory AT pingli anovelmoneylaunderingpredictionmodelbasedonadynamicgraphconvolutionalneuralnetworkandlongshorttermmemory AT feiwan novelmoneylaunderingpredictionmodelbasedonadynamicgraphconvolutionalneuralnetworkandlongshorttermmemory AT pingli novelmoneylaunderingpredictionmodelbasedonadynamicgraphconvolutionalneuralnetworkandlongshorttermmemory |