用于可疑金融交易监控的对比离群点检测模型

A Peer-group Outlier Detection Definition Model Applied to Suspicious Financial Transaction Surveillance

对基于单数据集和多数据集的离群点算法进行研究,提出一个基于距离模式进行数据集间参照对比的离群点判别模型,该模型通过数学定义清晰描述了参照集和对比集之间离群点模式的判别检测关系,为深入研究切合金融数据挖掘特点的算法建立形式化描述体系。这一模型也可推广应用于网络入侵检测、财务审计、图像识别、电子商务、医疗疫情监测等领域。

Outlier detection is a key element of financial surveillance systems which intend to identify credit card fraud, loan claim fraud and money laundering by discovering suspicious transaction. Financial outlier detection techniques generally fall into two categories, comparing every transaction against its account history and further more, comparing against a peer group to determine if the behavior is unusual, The later approach shows particular merits in efficiently extracting suspicious transaction and reducing false positive rate. Peer group analysis concept is largely dependent on a cross-datasets outlier detection model. The paper has intensively studied outlier detection among single data.set and cross datasets. An outlier detection model based on two peer datasets reference has been proposed incorporated with financial surveillance requests. The mathematical definition based on distance joined with local density gives an explicit the comparison procedure and discrimination criteria. It can be applied to IDS, auditing, health monitoring as well.