Global confidence degree based graph neural network for financial fraud detection
Graph Neural Networks (GNNs) are widely used in financial fraud detection due to their excellent ability on handling graph-structured financial data and modeling multilayer connections by aggregating information of neighbors. However, these GNN-based methods focus on extracting neighbor-level inform...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
ELS Publishing (ELSP)
2025-04-01
|
| Series: | Artificial Intelligence and Autonomous Systems |
| Subjects: | |
| Online Access: | https://elsp-homepage.oss-cn-hongkong.aliyuncs.compaper/journal/open/AIAS/2025/aias20250004.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849762673564057600 |
|---|---|
| author | Jiaxun Liu Yue Tian Guanjun Liu |
| author_facet | Jiaxun Liu Yue Tian Guanjun Liu |
| author_sort | Jiaxun Liu |
| collection | DOAJ |
| description | Graph Neural Networks (GNNs) are widely used in financial fraud detection due to their excellent ability on handling graph-structured financial data and modeling multilayer connections by aggregating information of neighbors. However, these GNN-based methods focus on extracting neighbor-level information but neglect a global perspective. This paper presents the concept and calculation formula of Global Confidence Degree (GCD) and thus designs GCD-based GNN (GCD-GNN) that can address the challenges of camouflage in fraudulent activities and thus can capture more global information. To obtain a precise GCD for each node, we use a multilayer perceptron to transform features and then the new features and the corresponding prototype are used to eliminate unnecessary information. The GCD of a node evaluates the typicality of the node and thus we can leverage GCD to generate attention values for message aggregation. This process is carried out through both the original GCD and its inverse, allowing us to capture both the typical neighbors with high GCD and the atypical ones with low GCD. Extensive experiments on two public datasets demonstrate that GCD-GNN outperforms state-of-the-art baselines, highlighting the effectiveness of GCD. We also design a lightweight GCD-GNN (GCD-GNNlight ) that also outperforms the baselines but is slightly weaker than GCD-GNN on fraud detection performance. However, GCD-GNNlight obviously outperforms GCD-GNN on convergence and inference speed. |
| format | Article |
| id | doaj-art-e40ee2a92fd84d039c426314af746d75 |
| institution | DOAJ |
| issn | 2959-0744 2959-0752 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | ELS Publishing (ELSP) |
| record_format | Article |
| series | Artificial Intelligence and Autonomous Systems |
| spelling | doaj-art-e40ee2a92fd84d039c426314af746d752025-08-20T03:05:41ZengELS Publishing (ELSP)Artificial Intelligence and Autonomous Systems2959-07442959-07522025-04-012112010.55092/aias202500041899084859445456896Global confidence degree based graph neural network for financial fraud detectionJiaxun Liu0Yue Tian1Guanjun Liu2The Department of Computer Science, Tongji University, Shanghai 201804, ChinaThe Department of Computer Science, Tongji University, Shanghai 201804, ChinaThe Department of Computer Science, Tongji University, Shanghai 201804, ChinaGraph Neural Networks (GNNs) are widely used in financial fraud detection due to their excellent ability on handling graph-structured financial data and modeling multilayer connections by aggregating information of neighbors. However, these GNN-based methods focus on extracting neighbor-level information but neglect a global perspective. This paper presents the concept and calculation formula of Global Confidence Degree (GCD) and thus designs GCD-based GNN (GCD-GNN) that can address the challenges of camouflage in fraudulent activities and thus can capture more global information. To obtain a precise GCD for each node, we use a multilayer perceptron to transform features and then the new features and the corresponding prototype are used to eliminate unnecessary information. The GCD of a node evaluates the typicality of the node and thus we can leverage GCD to generate attention values for message aggregation. This process is carried out through both the original GCD and its inverse, allowing us to capture both the typical neighbors with high GCD and the atypical ones with low GCD. Extensive experiments on two public datasets demonstrate that GCD-GNN outperforms state-of-the-art baselines, highlighting the effectiveness of GCD. We also design a lightweight GCD-GNN (GCD-GNNlight ) that also outperforms the baselines but is slightly weaker than GCD-GNN on fraud detection performance. However, GCD-GNNlight obviously outperforms GCD-GNN on convergence and inference speed.https://elsp-homepage.oss-cn-hongkong.aliyuncs.compaper/journal/open/AIAS/2025/aias20250004.pdfgraph neural networks (gnns)financial fraud detectionprototype learninggraph anomaly detection |
| spellingShingle | Jiaxun Liu Yue Tian Guanjun Liu Global confidence degree based graph neural network for financial fraud detection Artificial Intelligence and Autonomous Systems graph neural networks (gnns) financial fraud detection prototype learning graph anomaly detection |
| title | Global confidence degree based graph neural network for financial fraud detection |
| title_full | Global confidence degree based graph neural network for financial fraud detection |
| title_fullStr | Global confidence degree based graph neural network for financial fraud detection |
| title_full_unstemmed | Global confidence degree based graph neural network for financial fraud detection |
| title_short | Global confidence degree based graph neural network for financial fraud detection |
| title_sort | global confidence degree based graph neural network for financial fraud detection |
| topic | graph neural networks (gnns) financial fraud detection prototype learning graph anomaly detection |
| url | https://elsp-homepage.oss-cn-hongkong.aliyuncs.compaper/journal/open/AIAS/2025/aias20250004.pdf |
| work_keys_str_mv | AT jiaxunliu globalconfidencedegreebasedgraphneuralnetworkforfinancialfrauddetection AT yuetian globalconfidencedegreebasedgraphneuralnetworkforfinancialfrauddetection AT guanjunliu globalconfidencedegreebasedgraphneuralnetworkforfinancialfrauddetection |