Comprehensive data analysis on the EOS blockchain

Abstract

A blockchain is a growing list of blocks, or an open, distributed ledger, where anyone can record transactions and query logs, whereas the data cannot be altered retroactively once recorded. A consensus protocol of the blockchain concerns how to achieve an agreement among the nodes in the network. EOSIO (EOS) is one of the blockchain protocols that uses the Delegated Proof-of-Stake (DPoS) consensus algorithm and ranked in $9^{th}$ in the cryptocurrency market capitalization. However, there are a few possible flaws in the EOS blockchain. First of all, since an EOS stakeholder can vote for up to 30 candidates without any direct reward or cost, vote-trading can be a strong incentive. Moreover, since the voter participation rate and turnout are not enough, the large and active players can easily dominate votes even with the vote decay. Secondly, there can be wealth inequality because a large amount of token is stored on crypto exchanges or Dapp services and they can vote with these entrusted tokens. Lastly, there is no restriction for creating and running multiple accounts. In 2018, there was a shocking rumor concerning the major Chinese cryptocurrency exchange Huobi was accused of voter collusion in elections for block producers (BPs). However, there is a lack of research on such incidences. In this paper, we collect and analyze all of the EOS blockchain data for the first time, to the best of our knowledge. We perform flow analysis to detect and group multiple EOS account and find any signal of cheating on the BP election.