AI Insight
This paper presents a novel deep learning framework for detecting credit fraud at scale in Weixin Pay's digital payment system, which serves billions of users. The researchers developed a risk-aware method for training Graph Neural Networks on massive user-risk graphs by creating overlapping subgraphs that preserve critical fraud propagation patterns while maintaining computational efficiency. Their approach includes budget-constrained sampling to retain important long-tail evidence chains and a cross-subgraph alignment mechanism to ensure consistent user representations across distributed training partitions.
Why it matters
This work addresses a critical challenge in financial technology: detecting fraud among billions of users without losing important risk signals due to computational constraints. The framework could help digital payment platforms reduce financial losses while maintaining inclusive financial services for legitimate users, with potential applications across various large-scale fraud detection systems.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Abstract: Credit risk detection, particularly mitigating individual fraud, is crucial for maintaining the stability of digital financial ecosystems. Accurately identifying credit fraud among billions of users is critical for minimizing financial losses and safeguarding the sustainability of inclusive financial services. Given that credit fraud risks are often concealed within heterogeneous user-risk graphs, Graph Neural Networks (GNNs) have emerged as an effective tool for risk mining by capturing complex dependencies. To address the scalability bottleneck of industrial GNNs, distributed training based on subgraphs is indispensable. However, existing strategies often compromise topological integrity for load balancing. This can be catastrophic for risk detection, as it indiscriminately severs the long-tail evidence chains essential for risk propagation. Overlapping subgraphs can restore severed risk contexts but inevitably introduce redundancy and noise, while overlooking the representation alignment across different local subgraphs. In this paper, we propose a risk-aware overlapping subgraph learning framework for large-scale credit risk detection. We first construct base partitions to ensure load balance. Then, we perform budget-constrained sampling that selects informative long-tail nodes, thereby preserving critical risk diffusion patterns while filtering out noise. To mitigate representation inconsistency, we design a cross-subgraph consistency alignment mechanism. By enforcing alignment constraints on the overlapping nodes, we harmonize the local representations into a globally consistent latent space. Extensive experiments on Weixin Pay’s production dataset demonstrate that our model significantly outperforms existing strategies for risk detection, offering a scalable and effective solution for industrial graph learning.
Source: Empowering Credit Risk Detection in Weixin Pay with Billion-Scale Deep Graph Learning