Paper: arXiv 2306.04643

Authors: Mingxiao Song, Yunsong Liu, Agam Shah, Sudheer Chava

Abstract

The Non-Fungible-Token (NFT) market has experienced explosive growth in recent years. According to DappRadar, the total transaction volume on OpenSea, the largest NFT marketplace, reached 34.7 billion dollars in February 2023. However, the NFT market is mostly unregulated and there are significant concerns about money laundering, fraud and wash trading. The lack of industry-wide regulations, and the fact that amateur traders and retail investors comprise a significant fraction of the NFT market, make this market particularly vulnerable to fraudulent activities. Therefore it is essential to investigate and highlight the relevant risks involved in NFT trading. In this paper, we attempted to uncover common fraudulent behaviors such as wash trading that could mislead other traders. Using market data, we designed quantitative features from the network, monetary, and temporal perspectives that were fed into K-means clustering unsupervised learning algorithm to sort traders into groups. Lastly, we discussed the clustering results’ significance and how regulations can reduce undesired behaviors. Our work can potentially help regulators narrow down their search space for bad actors in the market as well as provide insights for amateur traders to protect themselves from unforeseen frauds.

Complexity vs Empirical Score

  • Math Complexity: 3.0/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The methodology relies on accessible K-means clustering and feature engineering rather than advanced mathematics, but the paper demonstrates strong empirical rigor through the use of real NFT market data (OpenSea, LooksRare), explicit feature definitions from network/monetary/temporal perspectives, and a clear path toward implementation for regulators and marketplaces.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Detect abnormal trading (e.g., wash trading)<br>to mitigate fraud risks in the NFT market"] --> B["Data Collection<br>Market data from NFT platforms<br>(OpenSea)"]
  B --> C["Feature Engineering<br>Quantitative features from:<br>- Network, Monetary, Temporal views"]
  C --> D["Clustering Analysis<br>K-means unsupervised learning<br>to group traders based on behavior"]
  D --> E["Results & Discussion<br>Identified high-risk trading groups<br>Recommended regulatory interventions"]
  E --> F["Outcomes & Impact<br>• Assists regulators in targeting bad actors<br>• Protects amateur traders from fraud"]

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 proposes a K‑means clustering approach using quantitative network, monetary and temporal features to identify abnormal NFT traders.

Source sentence
Using market data, we designed quantitative features from the network, monetary, and temporal perspectives that were fed into K-means clustering unsupervised learning algorithm to sort traders into groups.

Assumes 3

  • The transaction data collected from the blockchain is representative of overall NFT market activity. data
    Source sentence
    In total, we gathered over 1 million trans‑actions with 252,924 distinct wallets.
  • Clusters derived from K‑means correspond to meaningful trader behavior categories, including wash traders. model
    Source sentence
    Based on the seven clusters we derived from K-Means, five of them are marked with specific labels while the other two are potential wash trader candidates
  • Benford’s law and round‑number clustering are appropriate tools for detecting anomalies in NFT transaction prices. statistical
    Source sentence
    We first examined the first digit distribution of transaction prices against Benford’s law

Data

  • NFT transaction dataset · Ethereum blockchain crypto-onchain public
    NFT transactions across major marketplaces (e.g., OpenSea)
    Source sentence
    In total, we gathered over 1 million trans‑actions with 252,924 distinct wallets.
  • Ethereum‑USD and Gas‑to‑Ethereum exchange rates · Yahoo Finance alternative public
    Source sentence
    we also retrieved historical Ethereum to USD and Gas to Ethereum exchange rate data from Yahoo Finance.

Recognised sources: Yahoo Finance

Costs

Not applicable The paper reports no trading or execution result.

Finds 2

  • The wash trader percentage in the NFT market is estimated at 5.38% based on the clustering results. statistical
    Source sentence
    five of them are marked with specific labels while the other two are potential wash trader candidates, concluding that the wash trader percentage in the NFT market is 5.38%.
  • The first‑digit distribution of transaction prices deviates significantly from Benford’s law (Chi‑squared = 24688, rejecting the null at 5% confidence). statistical
    Source sentence
    the Chi-squared test yields a Chi value of 24688, which is significantly larger than the critical value. Therefore, the test rejects our null hypothesis at a confidence level of 5%.

Checked by others

Contradicted by

  • Can AI Detect Wash Trading? Evidence from NFTs (2023)
    On the question: What proportion of NFT trades involve wash‑trading?
    Compare the findings

    This paper: The wash trader percentage in the NFT market is estimated at 5.38% based on the clustering results.

    That paper: Approximately 38% of NFT trades and 60% of traded value across three major exchanges likely involve wash‑trading.

    A reports 5.38% wash traders; B reports ~38% of trades (and 60% of value) as wash‑trading, showing a large discrepancy.

  • NFT Wash Trading Detection (2023)
    On the question: What proportion of NFT market activity is attributable to wash trading?
    Compare the findings

    This paper: The wash trader percentage in the NFT market is estimated at 5.38% based on the clustering results.

    That paper: On average 0.136% of transactions, 0.157% of tokens, and 0.109% of wallets are involved in wash trading, generating total price manipulation, sales, and repurchase profit of around $900K, $1.1M, and -$1.6M respectively.

    Finding A reports ~0.11‑0.16% of transactions/tokens/wallets involved, while Finding B estimates 5.38% wash traders, showing a clear contradiction.

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.