Summary of Global Confidence Degree Based Graph Neural Network For Financial Fraud Detection, by Jiaxun Liu et al.
Global Confidence Degree Based Graph Neural Network for Financial Fraud Detection
by Jiaxun Liu, Yue Tian, Guanjun Liu
First submitted to arxiv on: 24 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Global Confidence Degree (GCD)-based Graph Neural Network (GCD-GNN) addresses the limitations of existing GNN-based methods in financial fraud detection by incorporating a global perspective. By calculating the GCD for each node, which evaluates its typicality, the model can generate attention values for message aggregation and capture both typical and atypical neighbors. This approach outperforms state-of-the-art baselines on two public datasets, demonstrating its effectiveness. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The new GCD-GNN model is designed to detect financial fraud by considering a global perspective in addition to neighbor-level information. It calculates the Global Confidence Degree (GCD) for each node, which helps identify typical and atypical neighbors. This approach shows promising results on public datasets, outperforming existing methods. |
Keywords
* Artificial intelligence * Attention * Gnn * Graph neural network
