Data Mining Model for Money Laundering Detection in Financial Domain

Abstract

One of the likeliest problems faced by the banks is the way the nature of money transaction takes place is not always known to the bank. One such problem consisting large amounts of money transferring through various accounts by the same person or entity is Money Laundering. Money laundering system is quite a convoluted process. It takes some understanding of the fund transferring activities at various stages. The system that works against Money laundering is Anti-Money Laundering (AML) system. The existing system for Anti-Money Laundering accepts bulk of data and converts it to large volumes reports that are tedious and topsy-turvy for a person to read without any help of software. Hence, the purpose of making decision within specified time span is de-facilitated. The basic motive behind designing a system is that it should either notify or generate alarm for Money Laundering process within time when the illicit actions are been carried out so that, some action can be taken or an decision could be made before the actual laundering task is completed. The Transaction flow analysis (TFA) is a system that facilitates the basic ideology of Anti-Money Laundering system. TFA consists of various segments those results in clusters of various sizes. However, the existing TFA system allows the data to be taken from a single entity may be bank or a financial institution. Also the cluster generated are large in volume hence the analyst sometime may find it difficult to analyze such huge amount of data. This project uses combination of TFA system with the system designed to extract customer behavior in order to make decision whether the customer is suspicious customer or not. Also this project will allow taking data as input from various sources, integrating and then performing operations on the data. General Terms Laundering