国内外反洗钱工作的大量实践表明,金融交易活动是洗钱犯罪行为的一个重要环节,通过分析金融机构的客户信息和交易数据,采用科学的方法识别可疑金融交易进而发现洗钱线索,已成为反洗钱研究的核心工作。文章将数据挖掘方法与金融领域知识相结合,首先通过对金融交易信息的多层次分析,总结出不同信息层次上的可疑金融交易特征;其次针对不同层次的交易信息,选择合适的数据挖掘方法,并结合客户背景资料,识别出可疑金融交易记录;最后依据贝叶斯判定原理,综合各层次的可疑信息,得到交易记录的整体可疑度,最终为反洗钱监测提供快速准确的参考。实验结果证明该方法是可行和有效的。
The practice of anti-money laundering at home and abroad indicates that financial transactions are the important step of money laundering crime. The core of anti-money laundering is to analyze the customer information and transaction data of financial organizations, recognize suspicious transactions based on scientific methods and find the clues of money laundering. Firstly, the paper summarizes the features of suspicious financial transactions through multi-level analysis of financial transaction information. Secondly, aiming at the transaction information with different levels, it recognizes the suspicious financial transaction records by using suitable data mining methods and customer information. At last, by Bayes' law and suspicious information with different levels, it obtains the whole suspicious degree of transaction records and provide references to the supervision over anti-money laundering. The empirical results provide support for the validity of the method above.