Hybrid LSTM-Graph Neural Framework for Robust Financial Fraud Detection and Adversarial Resilience

用于稳健金融欺诈检测与对抗性防御的混合 LSTM-图神经网络框架

Abstract: Financial institutions face significant challenges in detecting sophisticated money laundering patterns, such as smurfing and layering, due to extreme data imbalance (0.13% fraud rate) and evolving adversarial evasion tactics.

摘要： 由于极度的数据不平衡（欺诈率仅为 0.13%）以及不断演变的对抗性规避策略，金融机构在检测“拆分交易”（smurfing）和“分层洗钱”（layering）等复杂的洗钱模式时面临巨大挑战。

This paper proposes FraudShield AI, a hybrid framework that integrates Long Short-Term Memory (LSTM) networks with hand-crafted Graph Topological Features to capture both temporal sequences and structural relational context.

本文提出了 FraudShield AI，这是一个将长短期记忆（LSTM）网络与手工构建的图拓扑特征相结合的混合框架，旨在同时捕捉时间序列信息和结构化关系背景。

By engineering network-centric features including PageRank Centrality, In-Degree dynamics, and a custom Flow Ratio, the system shifts the detection paradigm from isolated transaction analysis to network-level forensics.

通过设计以网络为中心的特征（包括 PageRank 中心性、入度动态以及自定义的流量比率），该系统将检测范式从孤立的交易分析转向了网络层面的取证分析。

A Focal Loss objective is used to address class imbalance, and a dynamic thresholding mechanism is introduced to improve resilience against low-value smurfing attacks.

研究采用了焦点损失（Focal Loss）目标函数来解决类别不平衡问题，并引入了动态阈值机制，以提高系统对小额拆分交易攻击的防御能力。

Experimental evaluation on the PaySim dataset shows that the proposed hybrid model substantially outperforms Logistic Regression and XGBoost baselines in Precision, Recall, and F1-Score, particularly on hard-to-detect micro-transaction fraud patterns.

在 PaySim 数据集上的实验评估表明，该混合模型在精确率、召回率和 F1 分数方面均显著优于逻辑回归（Logistic Regression）和 XGBoost 基准模型，特别是在难以检测的微交易欺诈模式上表现突出。

An ablation study confirms the complementary contribution of both the temporal and topological components.

消融实验证实了时间组件和拓扑组件在模型中均发挥了互补性的贡献。