WTEYE: On-chain wash trade detection and quantification for ERC20 cryptocurrencies

Abstract

Wash trade is a common form of volume manipulation used to attract investors into the market and mislead them into making wrong investment judgments. Wash trade transactions are even more prominent in ERC20 cryptocurrencies. In this paper, we proposed two kinds of algorithms to reserve direct evidence of wash trade based on the on-chain transaction data of ERC20 cryptocurrencies. After labeling the wash trade, we continued to obtain features of the wash trade and quantify the volume of the wash trade. Our experiments show that for most ERC20 cryptocurrencies, the rate of wash trade reached over 15%. Specifically, over 30% of UNI token transactions ​were labeled as wash trade. It is demonstrated that the activations of most ERC20 cryptocurrencies are unreal, and restoring real data is necessary for market regulation.