Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics
Reviewed by Pithpith:YIDEPHHXopen to challenge →
read the original abstract
Anti-money laundering (AML) regulations play a critical role in safeguarding financial systems, but bear high costs for institutions and drive financial exclusion for those on the socioeconomic and international margins. The advent of cryptocurrency has introduced an intriguing paradox: pseudonymity allows criminals to hide in plain sight, but open data gives more power to investigators and enables the crowdsourcing of forensic analysis. Meanwhile advances in learning algorithms show great promise for the AML toolkit. In this workshop tutorial, we motivate the opportunity to reconcile the cause of safety with that of financial inclusion. We contribute the Elliptic Data Set, a time series graph of over 200K Bitcoin transactions (nodes), 234K directed payment flows (edges), and 166 node features, including ones based on non-public data; to our knowledge, this is the largest labelled transaction data set publicly available in any cryptocurrency. We share results from a binary classification task predicting illicit transactions using variations of Logistic Regression (LR), Random Forest (RF), Multilayer Perceptrons (MLP), and Graph Convolutional Networks (GCN), with GCN being of special interest as an emergent new method for capturing relational information. The results show the superiority of Random Forest (RF), but also invite algorithmic work to combine the respective powers of RF and graph methods. Lastly, we consider visualization for analysis and explainability, which is difficult given the size and dynamism of real-world transaction graphs, and we offer a simple prototype capable of navigating the graph and observing model performance on illicit activity over time. With this tutorial and data set, we hope to a) invite feedback in support of our ongoing inquiry, and b) inspire others to work on this societally important challenge.
This paper has not been read by Pith yet.
Forward citations
Cited by 14 Pith papers
-
SAGE: An LLM-driven Self Reflective Agentic Framework for Fraud Detection
SAGE is the first end-to-end LLM-driven multi-agent fraud detection system using a Data Diagnostic Tree and MDP optimization, achieving 40.86% average F1 gain and winning 96% of comparisons across five datasets and fi...
-
NeighborDiv: Training-free Zero-shot Generalist Graph Anomaly Detection via Neighbor Diversity
NeighborDiv detects graph anomalies via variance of inter-neighbor feature similarities under a new Neighbor-to-Neighbor Diversity Paradigm, achieving SOTA results with zero volatility in zero-shot cross-domain settings.
-
TRAVELFRAUDBENCH: A Configurable Evaluation Framework for GNN Fraud Ring Detection in Travel Networks
TravelFraudBench is a new configurable benchmark for GNN-based fraud ring detection in travel networks, simulating star, clique, and chain topologies and showing GraphSAGE outperforming MLP baselines on AUC and ring recovery.
-
MELT: A Behavioral Trace Dataset for High-Risk Memecoin Launch Detection
MELT is the first behavioral trace dataset for high-risk memecoin launch detection on Solana, providing 122 features, risk annotations, and ML benchmarks that reduce investment loss when used for selection.
-
Cross-Paradigm Graph Backdoor Attacks with Promptable Subgraph Triggers
CP-GBA distills a queryable repository of promptable subgraph triggers via graph prompt learning to achieve transferable backdoor attacks on GNNs with state-of-the-art success rates across paradigms and defenses.
-
TGHE: Template-based Graph Homomorphic Encryption for Privacy-Preserving GNN Inference in Edge-Cloud Systems
TGHE packs structurally identical ego-graph trees into shared CKKS ciphertexts for parallel encrypted GNN inference, delivering 66.9x speedup on a 3.7M-node financial graph with <0.002 AUC loss.
-
Provably Communication-Efficient and Privacy-Preserving Federated Graph Neural Networks
CE-FedGNN enables federated GNN training on coupled distributed graphs via infrequent aggregated representation exchange, moving-average estimation for staleness, and metric-DP, with O(1/sqrt(T)) convergence and O(T^{...
-
TransXion: A High-Fidelity Graph Benchmark for Realistic Anti-Money Laundering
TransXion supplies a 3-million-transaction graph benchmark with profile-aware normal activity and stochastic illicit subgraphs that produces lower detection scores than prior AML datasets.
-
TransXion: A High-Fidelity Graph Benchmark for Realistic Anti-Money Laundering
TransXion introduces a synthetic AML transaction benchmark with rich entity profiles and non-template adversarial anomaly synthesis that lowers detection model performance compared to prior benchmarks.
-
Agentic GraphRAG: Navigating Unstructured Financial Data with Collaborative AI
Agentic GraphRAG constructs a Neo4j graph via deterministic structured ingestion plus LLM extraction from notices, then deploys modular agents with tool access and reflection to outperform vector-RAG baselines on Swis...
-
Communication-free Sampling and 4D Hybrid Parallelism for Scalable Mini-batch GNN Training
ScaleGNN uses communication-free sampling and 4D parallelism to scale mini-batch GNN training to 2048 GPUs, achieving 3.5x speedup over prior state-of-the-art on ogbn-products.
-
Temporal Motif-aware Graph Test-time Adaptation for OOD Blockchain Anomaly Detection
TEMG-TTA combines temporal motif-aware graph learning with test-time adaptation to improve OOD anomaly detection on blockchain graphs, reporting an average 54.88% gain over prior GAD methods on five real-world datasets.
-
BlazingAML: High-Throughput Anti-Money Laundering (AML) via Multi-Stage Graph Mining
BlazingAML uses a multi-stage graph mining framework and compiler to express fuzzy AML patterns, matching SOTA F1 scores while delivering 210x CPU and 333x GPU speedups on IBM datasets.
-
Extracting Money Laundering Transactions from Quasi-Temporal Graph Representation
ExSTraQt uses quasi-temporal graph representations and supervised learning to detect suspicious transactions, achieving F1 score uplifts of up to 1% on real data and over 8% on synthetic datasets compared to prior AML models.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.