Change Address Detection in Bitcoin using Hierarchical Clustering
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The widespread adoption and success of blockchain, particularly Bitcoin, was influenced by the promise of decentralization and anonymity. Sadly, these same characteristics have made it attractive to illegal activities, requiring careful oversight and targeted interventions. In order to mitigate illicit usage of this technology, we need to analyze and de-anonymize transactions occurring in the blockchain. For that purpose, change addresses identification is a promising technique, since change addresses can be associated to the inputs of the same transaction since they are meant to hold leftover funds for the same user. In this article, we propose a new approach of change address detection using hierarchical clustering. First, we developed a new method for data extraction of connected transactions. After collecting the transaction, we combined multiple input heuristics with a hierarchical clustering algorithm at the transaction level to study similarities in usage patterns between inputs and outputs. After applying our detection model, we analyze the generating cluster and evaluate the performance of our solution in terms of F1-score, result accuracy, recall and precision.
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.
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 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| 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 it