Financial transaction network method improves detection of illicit accounts
MINT: Mitigating Structural Inconsistency via Attentive Message Passing for Financial Anomaly Detection
Machine Learning
Summary
Detecting illegal activity in financial networks is hard because criminals try to hide by creating misleading connections, and there are usually very few known examples of bad actors to learn from. The authors propose MINT, a new method that looks carefully at how money flows back and forth between accounts and adjusts how strongly accounts are connected based on suspicious patterns. It also uses special learning techniques that work well even when only a few accounts are labeled as illegal. Their tests on multiple financial datasets show MINT finds bad accounts more effectively than existing methods.
What this means in practice
- •For financial fraud detection teams: Enhance identification of suspicious accounts by applying attentive message-passing to transaction data.
- •For cybersecurity operations: Improve detection of deceptive network behavior in monetary transactions by dynamically adjusting connection importance.
Authors
Yin-Ju Wu, Bi-Cheng Yan, Berlin Chen
Abstract
Illicit account detection is a critical task for financial risk control systems, aiming to pinpoint malicious actors within transaction networks. Existing efforts typically recast transaction networks into directed multigraphs and leverage graph neural networks (GNNs) to detect anomalies via topological structures under the homophily assumption. However, directly grafting GNNs onto transaction graphs is faced with at least two pivotal challenges, i.e., structural inconsistency and label scarcity. The former arises from deceptive connections forged by malicious actors, while the latter stems from the lack of annotated data. In light of these, we introduce MINT, a novel graph neural framework that mitigates structural inconsistency via an attentive message-passing scheme on transaction graphs. Specifically, MINT introduces a discrepancy-aware feature propagation, which tracks malicious actors by quantifying deviations between nodes across bi-directional transaction flows. A transaction salience module is proposed to mitigate structural inconsistency by adaptively regularizing the interaction strength between nodes. Furthermore, to alleviate label scarcity, MINT is optimized via Positive-Unlabeled (PU) learning, which exploits only a minority of illicit nodes coupled with a majority of unlabeled ones. Extensive experiments conducted on three financial transaction benchmarks, including two large-scale public and one in-house real-world datasets, validate the superior effectiveness of MINT over several cutting-edge baselines.