solsticestudioai/dark-pool-pack

Aestrea Dark Pool金融异常包（样本）是一个合成的金融异常和欺诈检测数据集，用于AML/交易监控模型训练、误报减少研究以及金融科技异常检测流程。每一行都是一个完整的金融活动序列——一个休眠的预热期后跟随一个突然的活动爆发——标记为账户接管、市场操纵欺骗或合法但高风险（误报）行为，包含遥测、身份背景、检测逻辑和金融暴露。该数据集由SolsticeAI构建，作为更大商业包的免费样本，100%合成，不包含真实交易、账户、地址、订单簿或客户数据。数据集包含10,000个金融活动序列，分为3种欺诈类别，并覆盖多种资产和地理区域。

The Aestrea Dark Pool Financial Anomaly Pack (Sample) is a synthetic financial-anomaly and fraud-detection dataset for AML / transaction-monitoring model training, false-positive reduction research, and fintech anomaly-detection pipelines. Each row is a complete financial activity sequence — a dormant warm-up period followed by a sudden activity burst — labeled as account takeover, market-manipulation spoofing, or legitimate-but-high-risk (false positive) behavior, with telemetry, identity context, detection logic, and financial exposure. Built by SolsticeAI as a free sample of a larger commercial pack, it is 100% synthetic with no real transactions, accounts, addresses, order books, or customer data. The dataset includes 10,000 financial activity sequences, stratified into 3 fraud classes, and covers multiple assets and geographic regions.