Paper: arXiv 2411.13603

Authors: Ling Cheng, Qian Shao, Fengzhu Zeng, Feida Zhu

Abstract

Since its advent in 2009, Bitcoin (BTC) has garnered increasing attention from both academia and industry. However, due to the massive transaction volume, no systematic study has quantitatively measured the asset decentralization degree specifically from a network perspective. In this paper, by conducting a thorough analysis of the BTC transaction network, we first address the significant gap in the availability of full-history BTC graph and network property dataset, which spans over 15 years from the genesis block (1st March, 2009) to the 845651-th block (29, May 2024). We then present the first systematic investigation to profile BTC’s asset decentralization and design several decentralization degrees for quantification. Through extensive experiments, we emphasize the significant role of network properties and our network-based decentralization degree in enhancing Bitcoin analysis. Our findings demonstrate the importance of our comprehensive dataset and analysis in advancing research on Bitcoin’s transaction dynamics and decentralization, providing valuable insights into the network’s structure and its implications.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 8.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper centers on a massive, real-world 15-year dataset and practical graph analysis techniques like community detection and centrality metrics, indicating high empirical rigor. The math is moderate, involving network science and statistical definitions of centrality, but lacks heavy theoretical derivations or complex modeling.

Research Flowchart

  flowchart TD
  Start["Research Goal: Quantify BTC Asset Decentralization<br>via Network Properties"] --> Input["Data Input:<br>Full-history BTC Transaction Network<br>(15 Years, 2009-2024)"]
  Input --> Process["Methodology:<br>1. Graph Construction & Analysis<br>2. Define Network-based Decentralization Metrics"]
  Process --> Compute["Computational Analysis:<br>Structural Property Extraction<br>Quantification of Decentralization Degrees"]
  Compute --> Outcomes["Key Findings:<br>- Created first systematic full-history dataset<br>- Network metrics reveal structural shifts<br>- New decentralization degrees provide robust quantification"]
  Outcomes --> Impact["Outcome:<br>Foundation for advanced Bitcoin research<br>and dynamic asset profiling"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper introduces the first full‑history Bitcoin transaction network dataset covering the genesis block to block 845,651 and proposes a network‑based decentralization degree (DHHI) to quantify Bitcoin asset decentralization.

Source sentence
Our contributions can be summarized as follows: • We introduce the first full-history BTC transaction net-work dataset, covering from the genesis block to block 845651 (May 29, 2024). We thus propose a decentralization degree DHHI based on the A1 and A2 clustering approach, A higher DHHI indicates greater decentralization.

Assumes

No explicit assumptions found in the text.

Data

  • full‑history BTC transaction network dataset · BTC Core with BlockSci crypto-onchain public
    2009-2024 · all Bitcoin addresses
    Source sentence
    Our contributions can be summarized as follows: • We introduce the first full-history BTC transaction net-work dataset, covering from the genesis block to block 845651 (May 29, 2024). B. Raw Data Our dataset spans from January 4, 2009, to May 29, 2024, covering more than 15 years of BTC transactions.

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • iTransformer consistently achieves the lowest prediction errors across all historical lengths and horizons for transaction fee forecasting. predictive-accuracy
    Source sentence
    For 1-day predictions, iTransformer consistently achieves the lowest errors across all historical lengths, while other models show higher errors.
  • For 7‑day fee predictions, iTransformer attains MSE 0.354 and MAE 0.194, outperforming other models. performance
    Source sentence
    For 7-day predictions, errors increase but iTransformer remains the best, particularly for 7-to-7 predic-tions in transaction fees (MSE: 0.354, MAE: 0.194) and 30-to-7 in MVRV-Z (MSE: 0.010).
  • Adding Asset and Rank features to Centrality reduces iTransformer’s MSE for transaction fee prediction from 0.474 to 0.435. predictive-accuracy
    Source sentence
    III and Fig. 6), adding Asset to Centrality reduces the MSE for iTransformer from 0.474 to 0.456, and further decreases to 0.435 when Rank is also added.
  • In MVRV‑Z prediction, incorporating Asset and Rank features improves model performance across most models. predictive-accuracy
    Source sentence
    V and Fig. 7), a similar trend is seen, with performance improving across most models when Asset and Rank features are added.
  • The proposed DHHI metric, based on address‑level or clustered entities, provides a quantifiable measure where higher values indicate greater decentralization. theoretical
    Source sentence
    We thus propose a decentralization degree DHHI based on the A1 and A2 clustering approach, A higher DHHI indicates greater decentralization.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 27 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (48)

  1. On scaling decentralized blockchains (2016)
  2. Decentralization in bitcoin and ethereum networks (2018)
  3. Is bitcoin a decentralized currency? (2014)
  4. Quantitative analysis of the full bitcoin transaction graph (2013)
  5. Trend of centralization in Bitcoin’s distributed network (2015)
  6. Quantifying Decentralization (2017)
  7. An Information Entropy Method to Quantify the Degrees of Decentralization for Blockchain Systems (2019)
  8. A Coefficient of Variation Method to Measure the Extents of Decentralization for Bitcoin and Ethereum Networks. (2020)
and 40 more
  1. Comparison of Decentralization in DPoS and PoW Blockchains (2020) — link
  2. Do the rich get richer? An empirical analysis of the Bitcoin transaction network (2014)
  3. Empirical analysis of crypto currencies (2016)
  4. The bow tie structure of the bitcoin users graph (2019)
  5. BlockSci: Design and applications of a blockchain analysis platform (2020)
  6. Bitcoin: A peer-to-peer electronic cash system (2019)
  7. Sok: Research perspectives and challenges for bitcoin and cryptocurrencies (2015)
  8. Making smart contracts smarter (2016)
  9. Decentralized consensus for edge-centric internet of things: A review, taxonomy, and research issues (2017)
  10. Pervasive decentralisation of digital infrastructures: a framework for blockchain enabled system and use case analysis (2017)
  11. Securing virtual machine orchestration with blockchains (2017)
  12. A survey on consensus mechanisms and mining strategy management in blockchain networks (2019)
  13. Forecasting Bitcoin price with graph chainlets (2018)
  14. Blockchain: A graph primer (2017) — link
  15. Bitcoin: Economics, technology, and governance (2015)
  16. Decentralization is vulnerable under the Gap game (2019)
  17. Bitcoin and the myth of decentralization: Socio-technical proposals for restoring network integrity (2016)
  18. Peer to peer for privacy and decentralization in the internet of things (2017)
  19. Detecting Ponzi Schemes on Ethereum: Towards Healthier Blockchain Technology (2018)
  20. Measurements, Analyses, and Insights on the Entire Ethereum Blockchain Network (2020)
  21. Blockchain Mining Games with Pay Forward (2019)
  22. Identifying systemically important financial institutions: a network approach (2019)
  23. Trade Credit, Financial Intermediary Development, and Industry Growth (2003)
  24. How Unstable Are Complex Financial Systems? Analyzing an Inter-bank Network of Credit Relations (2013)
  25. Similarities and differences between the CR and HHI as an indicator of market concentration and market power (2016)
  26. Concentration in US labor markets: Evidence from online vacancy data (2020)
  27. SNAP Datasets : Stanford Large Network Dataset Collection (2014)
  28. Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics (2019) — link
  29. Predicting the price of Bitcoin by the most frequent edges of its transaction network (2019) — link
  30. Bitcoin risk modeling with blockchain graphs (2018) — link
  31. Analysis of cryptocurrency transactions from a network perspective: An overview (2021)
  32. Blockchain Data Mining With Graph Learning: A Survey (2023)
  33. Attention is all you need (2017)
  34. Reformer: The Efficient Transformer (2019)
  35. Informer: Beyond efficient transformer for long sequence time-series forecasting (2021)
  36. Flowformer: Linearizing Transformers with Conservation Flows (2022)
  37. Flashattention: Fast and memory-efficient exact attention with io-awareness (2022)
  38. Learning phrase representations using RNN encoder-decoder for statistical machine translation (2014) — link
  39. itransformer: Inverted transformers are effective for time series forecasting (2023) — link
  40. Long Short-term Memory (1997)

0 of 48 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.