9.2 Money As Data


9.2 Money As Data

In this era of interconnected networks, money represents information moving at the speed of light, where digital crime within the banking and financial services industry can cost billions of dollars each year—undetected and unreported. Using more sophisticated detection policies and procedures has become a necessity, and these include the use of investigative data mining. While some detection techniques are common to all frauds, the investigative data miner must understand the different nature of financial crimes in order to develop a methodology for uncovering them. Financial fraud often consists of repeated criminal incidents involving many transactions using methods of operation similar in content and appearance, but not quite identical.

Financial institutions should not rely solely on the experience and intuition of auditors, analysts, and fraud specialists to detect criminal transactions. In most cases, there is a history of audited fraudulent transactions that can be used to build models to predict future fraudulent activities. The goal of data mining for a financial institution is the development of rules and models enabling it to reduce the number of fraudulent-transaction alerts to a volume that can be handled and investigated by, say, an audit group in a bank. The problem is that the perpetrators of the crimes can learn the rules and procedures adopted by that audit group and take detection-avoidance action, such as changing their methods of operation, profiles, and dollar thresholds.

An effective analytic solution needs to capture the experience of fraud specialists, as well as incorporate the rules from analysis and predictive models. Together, a hybrid mechanism can be established in a flexible and sophisticated manner that makes avoidance by criminals difficult. We will describe a number of fraud scenarios and indicate similarities and some known signatures. It is very important to learn as much as possible about the nature of the crimes and the criminals' MOs. Fraud detection is a process; it never stops. It just evolves over time with new schemes, stings, scams, and MOs, but so do the weapons for detecting and deterring them.




Investigative Data Mining for Security and Criminal Detection
Investigative Data Mining for Security and Criminal Detection
ISBN: 0750676132
EAN: 2147483647
Year: 2005
Pages: 232
Authors: Jesus Mena

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net