Research On Recognition Of Suspicious Transactions Based On Outlier Analysis

Money laundering is increasing dramatically with the expansion of modern technology and the global superhighways of communication, resulting in the loss of billions of dollars worldwide each year. Money laundering is an economically significant crime. Several hundred billion dollars are washed through the financial sector in the world, and money laundering facilitates crimes as harmful as drug trafficking and terrorism. Choosing money laundering enforcement as the leading example is motivated by the fact that the identification role of report is particularly strong. The bank monitors transactions and reports suspicious activity to the government which identifies targets for investigation based on these reports. However excessive reporting fails to identify what is truly important by diluting the information value of reports. The reporting problem is investigated through the first analysis of money laundering enforcement.Outlier analysis is an important part of data mining. Its purpose is to find the "small patterns" from dataset. An outlier is an object that is considerably dissimilar or inconsistent with the remainder of the data. This is very useful in anti money laundering. It is important to find them quickly and accurately. The outlier detection technology adapted to the recognition of suspicious transaction is discussed in this thesis.Data mining and machine learning provide effective technologies for money laundering detection and have been applied successfully to detect money laundering activities. We describe the tools available for money laundering detection and the areas in which money laundering detection technologies are most used. Based on the analysis of current anti-money laundering in China, research was conducted on the method to recognize the money laundering via data-mining. In this process, outlier detection technology, grid and tense-based clustering algorithm have been applied.