一种高维聚类算法及在洗钱侦测中的应用

A High Dimension Clustering Algorithm and its Application in Detecting Money Laundering

从技术的角度看,洗钱侦测问题实际上是一个数据分析问题。本文首先给出了一个可疑交易判定模型,并提出了一个基于超图模型的高维聚类算法,运用该算法从案例库中形成可疑交易模式,最后给出了可疑交易的判定方法。该基于超图的高维聚类算法具有以下特点:1)能处理大数据集;2)能适应高维数据;3)聚类结果是可理解、可解释和可用的。

From the point of view of technique, detectiing money laundering is process of data analysis. This paper first proposes a hypergragh-based clustering algorithm to find out suspicious business mode about money laundering. And then it discusses way of determination suspicious business. The algorithm could solve the problems of 1）large volume of data set; 2）data set of high dimension; 3）easy understanding of result. The algorithm has been employed in the money laundering detection system, which is the first anti-money laundering system of our nation.