Advanced Financial Fraud Detection Using GNN-CL Model
By Yu Cheng, Junjie Guo, Shiqing Long, You Wu, Mengfang Sun, Rong Zhang
Rating
825
Battle Count: 152
Relevance
6/10
While focused on fraud detection, the techniques could be adapted for detecting anomalies or unusual patterns in trading data.
Implementation Complexity
7/10
The model combines multiple complex components (GNN, CNN, LSTM) and includes custom modules like the neighborhood noise purifier and core node intensifier, which may require significant effort to implement correctly.
Reproducibility
3/5
The paper provides some details on the model architecture and experimental setup, but lacks specific hyperparameters and full implementation details.
About this paper
Methodology: GNN-CL. Problem types: Classification, Anomaly Detection.
The interactive Everscope explorer (charts, battles, favorites) loads below.