AI-Based Framework for Classifying Cryptocurrency Exchanges through Transaction Feature Analysis
Bibliographic record
Abstract
Money laundering through cryptocurrencies has been increasing in recent years. To prevent the misuse of virtual assets for money laundering, we perform exchange classification. In this paper, we introduce an AI-based framework that automatically classifies cryptocurrency exchanges by utilizing wallet addresses and transaction data. Specifically, we collected approximately 2 million wallet addresses and transaction data for 50 exchanges. To address the limitations of single models and maximize classification performance, we applied Graph AI models, including the Heterogeneous Graph Attention Network and the Relational Graph Convolutional Network, in combination with ensemble techniques. As a result, our proposed frameworks achieved an average F1-score of 0.8676 across the 50 exchanges. Furthermore, this framework significantly improves the accuracy of tracking wallets involved in cryptocurrency crimes and fosters stronger cooperation with exchanges. The key contributions of this study include the following: 1) Tackling the understudied problem of exchange classification for anti-money laundering, 2) Developing a high-quality labeled dataset of exchange wallet addresses and transactions, and 3) Designing an effective feature set and graph AI approach for exchange classification based on transaction pattern analysis. Ultimately, our extensive experimental results demonstrate the potential for improving cryptocurrency crime detection and recovery through advanced AI techniques.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".