Recognizing On Suspicious Money Laundering Transactional Activities Based On Corporate Bank Accounts

The crime of Money Laundering (ML) can be an independent criminal behavior, and can also be predicate crime and lower crime of other criminal activities. Therefore, ML will not only have an impact on the country's economic development, but also will encourage criminals, threaten national stability and do harm to nations' images. With the development of economic globalization and international financial markets, Anti-money Laundering (AML) becomes an important topic of international financial field. Meanwhile, Suspicious Activity Report (SAR) system is considered as the foundation of Anti-money Laundering. Both in FinCen and China Anti-Money Laundering Monitoring & Analyzing Center, the Large and Suspicious Transactions used for analyzing are supplied by SAR system. The current SAR system adopted by China has such limitations as fuzzy criterions, less of adaptive abilities, easy to avoid and massive data, etc. The main purpose of this paper is to put forward an automatic and dynamic suspicious transactional activities recognition method which is difficult to avoid and easy to measure. All of these are based on current SAR system and bank accounts. This is rather significant for the development of SAR system and accord with the future research direction of AML.The up-to-minute research results of suspicious transaction activities recognition at home and abroad are greatly accepted and developed in this thesis. Base on the analysis of current SAR system in China, it does research on suspicious activities recognition via data mining. In this process, Outlier Detection Technology, Grid-based clustering algorithm and LOF algorithm have been applied.Firstly, the current AML procedures in China are analyzed. At the same...