Elliptic2
收藏资源简介:
Elliptic2数据集由麻省理工学院计算机科学与人工智能实验室与Elliptic合作创建,包含121,810个标记的比特币集群子图,位于一个包含4900万个节点集群和1.96亿次边缘交易的背景图中。该数据集专注于通过子图表示学习揭示加密货币中的洗钱活动模式,并准确分类新的犯罪活动。创建过程中,采用了多种技术和专业知识来标记和分析子图。Elliptic2数据集的应用领域主要集中在反洗钱和加密货币的司法分析,旨在通过机器学习技术提高对非法金融活动的识别能力。
The Elliptic2 dataset was developed jointly by the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) and Elliptic. It consists of 121,810 labeled Bitcoin cluster subgraphs, which are embedded in a background graph containing 49 million node clusters and 196 million transaction edges. This dataset aims to uncover money laundering patterns in cryptocurrencies through subgraph representation learning and accurately classify emerging criminal activities. Multiple technologies and professional expertise were utilized during its creation to label and analyze the subgraphs. The main application fields of the Elliptic2 dataset focus on anti-money laundering (AML) and cryptocurrency forensic analysis, with the objective of improving the identification of illegal financial activities via machine learning techniques.

- 1The Shape of Money Laundering: Subgraph Representation Learning on the Blockchain with the Elliptic2 Dataset麻省理工学院计算机科学与人工智能实验室 · 2024年



