EX-Graph
收藏资源简介:
EX-Graph是由新加坡国立大学创建的一个创新性开源数据集,首次真实地连接了以太坊和X平台。该数据集结合了以太坊交易记录(200万个节点和3000万条边)以及X平台的关注数据(100万个节点和300万条边),通过OpenSea验证了30,667个以太坊地址与X账户的关联。EX-Graph不仅捕捉了区块链上的金融交易,还涵盖了X平台上关于以太坊的社交互动,提供了一个更为全面的以太坊视图。该数据集通过详细的统计分析和实验,证明了通过验证匹配链接整合X数据显著提升了以太坊任务的性能,不仅促进了区块链研究,还推动了图学习的进步。EX-Graph的应用领域包括以太坊链接预测、洗钱交易地址检测以及以太坊与X账户匹配链接的预测,旨在解决区块链分析中的匿名性和数据限制问题。
EX-Graph is an innovative open-source dataset developed by the National University of Singapore, which for the first time establishes a realistic connection between the Ethereum blockchain and the X platform. This dataset integrates Ethereum transaction records (2 million nodes and 30 million edges) and X platform follow relationship data (1 million nodes and 3 million edges), and uses OpenSea to verify the association between 30,667 Ethereum addresses and X accounts. EX-Graph not only captures on-chain financial transactions but also covers Ethereum-related social interactions on the X platform, delivering a more holistic view of the Ethereum ecosystem. Through rigorous statistical analyses and controlled experiments, this dataset demonstrates that integrating X platform data via verified matching links significantly improves the performance of Ethereum-related tasks, which not only advances blockchain research but also promotes the development of graph learning. The application scenarios of EX-Graph include Ethereum link prediction, detection of money-laundering transaction addresses, and prediction of matching links between Ethereum addresses and X accounts, aiming to address the challenges of anonymity and data limitations in blockchain analytics.




