Soft Computing Tools in Credit card fraud & Detection Rashmi G.Dukhi G.H.Raisoni Institute of Information & Technology, Nagpur rashmidukhi25@gmail.

Transcription

1 Soft Computing Tools in Credit card fraud & Detection Rashmi G.Dukhi G.H.Raisoni Institute of Information & Technology, Nagpur Abstract Fraud is one of the major ethical issues in the credit card industry. The significance of the application of the techniques reviewed here is in the minimization of credit card fraud. Depending on the type of fraud faced by banks or credit card companies, various measures can be adopted and implemented. Neural Network is used to describe a statistical model that evaluates plastic card account and/or transaction data.the Neural Network Model generates a score for each account/transaction. The score separates those accounts/ transactions that are likely to be fraudulent from those that are not. Accounts/transactions that receive higher scores are more likely to be fraudulent.accounts/transactions that receive higher scores are more likely to be fraudelent. Keywords Credit card fraud, Detection techniques, Fraudster,Neural Network,Skimming I. INTRODUCTION Fraud means obtaining services/goods and/or money by unethical means, and is a growing problem all over the world nowadays. Fraud deals with cases involving criminal purposes that, mostly, are difficult to identify. Credit cards are one of the most famous targets of fraud. Credit card fraud may happen in various ways, which depend on the type of fraud concerned; it encapsulates bankruptcy fraud, theft fraud / counterfeit fraud, application fraud and behavioral fraud. Each of these sub-fraud categories has its own definition and specificity.section II describes the various types of frauds.section III explains how neural network can be applied for detecting fraudulent transactions. A. Perception of Risk One reason these numbers remain low is the perception of risk associated with the issuance of debit cards. This risk can be broken down into three categories: 1. Lost/stolen - Crime of opportunity, fraud is usually in small amounts and limited to one account 3. Friendly fraud - Fraud perpetrated by cardholders through system glitches, closing accounts before POS items clear and/or charge backs for valid transactions B. New Fraud Trends It is clear that issuers, especially smaller issuers, simply can't afford to be complacent. Often, criminals prey on institutions that they feel are less sophisticated in their risk management. New trends have emerged and are becoming more widespread in counterfeit fraud[10]. A few of the new methods that criminals are using to perpetuate fraud are: 1.Skimming - Criminals take valid cards and swipe them through a device that is about the size of a pager. The card information is then transferred and encoded onto any card with a magnetic stripe. The card could be a credit card that was recently stolen or found in the trash, any type of loyalty card or even a blank plastic card that can be purchased at a variety of places with no questions asked. The plastic is then used to perform transactions at locations where consumers swipe their own cards, such as gas stations, pay phones, and merchants where the customer passes their card through the POS terminal. Due to the fact that all of the valid track 2 data is reproduced, this type of fraud corrupts the CVV/CVC protection that card associations implemented just a few years ago[3]. ATM skimming is a type of fraud and figure 1 shows how PIN is captured by flashing light. 2.Account Generation Software - This type of software, available on the Internet, is used by criminals to obtain valid BINs and uses the proper algorithms to generate valid 16 digit account numbers that may or may not be issued. Criminals then take all of those valid account numbers and via computer or unmanned point of sale locations, start seeking vulnerable BIN. 2. Counterfeit - Process in which cards or accounts are created by criminals who have no association with the bank or cardholder 60

2 In other words, purchasers use credit cards knowing that they are not able to pay for their purchases. The bank will send them an order to pay. However, the customers will be recognized as being in a state of personal bankruptcy and not able to recover their debts. The bank will have to cover the losses itself. Usually, this type of fraud loss is not included in the calculation of the fraud loss provision as it is considered a chargeoff loss. The only way to prevent this bankruptcy fraud is by doing a pre-check with credit bureaux in order to be informed about the banking history of the customers. 2. Theft fraud/counterfeit fraud- Fig 1.An ATM skimming & Pin Capturing Device. The flashing light is easily observed. 3. Pinging - A method in which criminals try to get an authorization on the account numbers they have generated with account generation software. Once an authorization is given, multiple cards are produced and passed out to other members involved in the fraud ring. 4. Account Takeover - This type of fraud is becoming more prevalent. The criminal calls the bank to change the address on an account and then calls back a few days later explaining that their card was lost and requests a new debit card. Many technology controls and tools are being used today by financial institutions including card activation (cards are mailed inactive and can only be activated by VRU or PIN-based transactions at the ATM), address verification in non face-to-face transactions, CVC/CVV validation and expiration date checking[13]. With these tools in place, most instances of fraud can be avoided. However, it's the implementation of neural networks that has had the strongest effect on the reduction of fraud. 1. Bankruptcy fraud II. TYPES OF FRAUD Bankruptcy fraud is one of the most difficult types of fraud to predict. However, some methods or techniques may help in its prevention. Bankruptcy fraud means using a credit card while being insolvent. 61 Theft fraud means using a card that is not yours. The perpetrator will steal the card of someone else and use it as many times as possible before the card is blocked. The sooner the owner will react and contact the bank, the faster the bank will take measures to stop the thief. Counterfeit fraud occurs when the credit card is used remotely; only the credit card details are needed. At one point, one will copy your card number and codes and use it via certain web-sites, where no signature or physical cards are required On-line merchants are at risk because they have to offer their clients payment by credit card[5]. In cases where fraudsters use stolen or manipulated credit card data the merchant loses money because of so-called "charge-backs"2. Note that charge-backs are generated if credit card holders object to items on their monthly credit card statements because they were not responsible for the purchase transactions. A fraudulent transaction is difficult to detect and to define. Nevertheless, ATM transactions of large amounts are suspicious and demand contact with the customer. Purchases of goods for a larger amount than normal will also be notified to the customer as well as abnormal overseas spending patterns.fraudulent transactions are usually impossible to prevent as they occur in a really short period of time. However, once a card is identified, the card is blocked. 3. Application Fraud Application fraud is when someone applies for a credit card with false information. To detect application fraud, the solution is to implement a fraud system that allows identifying suspicious applications.

3 To detect application fraud, two different situations have to be distinguished: when applications come from a same individual with duplicates, and when applications come from different individuals with similar details, the so-called identity fraudsters.in most banks, to be eligible for a credit card, applicants need to complete an application form. This application form is mandatory except for social fields. The information required includes identification information, location information, contact information, confidential information and additional information. Recurrent information available would be for identification purposes, such as the full name and the date of birth. The applicant would inform the bank about his/her location details: the address, the postal code, the city and the country. The bank would also ask for contact details, such as address, land-line and mobile phone numbers. Confidential information will be the password. In addition, the gender will be given. All those characteristics may be used while searching for duplicates. To identify the so-called duplicates, cross-matching techniques are in common use. Rather than using statistical techniques, another method easy to implement is cross-matching. For instance, simple queries that give fast results are to cross-identify information with location details. Examples would be last name and date of birth and postal code and address or last name and address and and gender. By those queries, individuals with more than one card are identified. Those are quite simplistic queries but will remove most duplicates from the system. Note that duplicates may usually be genuine. Customers can reapply filling in a new address or spelling differently in one of the fields. By contrast, identity crime, as it is named, is perpetrated by real criminals filling wrong application data consciously. 4. Behavioral fraud - Behavioral fraud occurs when details of legitimate cards have been obtained\ fraudulently and sales are made on a cardholder present basis. These sales include telephone sales and e-commerce transactions, where only the card details are required.behavioral fraud can be detected by implementing a fraud scorecard predicting which customers are likely to default. Traditional credit scorecards are used to detect customers who are likely to default, and the reasons for this may include fraud.regarding the process, using scoring for fraud prevention is similar to any other use, including profit, default, and collection. The score reflects experience of past cases, and the result is a binary outcome: a genuine customer or a fraudster. III. NEURAL NETWORK APPROACH Neural networks are extremely important in today's banking world and provide many benefits. Unlike daily reports, all suspect activity is detected immediately and can be tied to the authorization system to deny after the first detection. In addition, neural networks are always adapting to new fraud trends. Criminals can by-pass all of the fraud methods discussed in some manner; however, neural networks are the only systems that help detect unusual activity when all other methods of fraud prevention have been circumvented. "Self-learning" software that builds and learns the characteristics of each cardholder can "predict" if a transaction may be fraudulent. The term for this type of fraud control is Predictive Modeling. This software takes every transaction and provides a score based on previous activity for that account. If the transaction has some element (amount, geography, merchant type, etc.) that causes the software to "predict" that it is probably fraudulent, it gives it a very high score. These neural networks can be installed in tandem with the bank processor's authorization system, so that transactions are scored immediately. Based on the score received, they can be further investigated or declined. In the world of plastic card fraud, the term Neural Network is used to describe a statistical model that evaluates plastic card account and/or transaction data. Typically, the Neural Network Model generates a score for each account/transaction. The score separates those accounts/ transactions that are likely to be fraudulent from those that are not. Accounts/transactions that receive higher scores are more likely to be fraudulent. A robust Neural Network Model helps a fraud investigator focus on those accounts/transactions that are MOST LIKELY to be fraudulent [14]. 62

4 A Neural Network Model is one component of a more complex Fraud Detection System. There are many other tools that are used to deliver and augment the score created by the Neural Network Model. These include: A. Rule Systems allow issuers and processors to focus more specifically on fraudulent accounts/transactions. Rule Systems typically incorporate a Neural Network Model score and allow the issuer to apply their own detailed knowledge to improve the process. For example: If an issuer that typically reviews cases with a Neural Network score of 800 suddenly experiences an outbreak of fraud at electronic stores in Madison, WI, a Rule System would allow them to create a rule that says: IF the MERCHANT is ELECTRONICS AND the zip code is between and AND the Neural Network Score is greater than 600 CREATE A CASE TO REVIEW In this case, the rule allows the issuer to focus on a specific fraud situation that would have been missed using the Neural Network score of 800. The important distinction is that rules can be modified by the user, while the Neural Network Model cannot. B. Case Management Systems- refer to the PC workstation that a fraud analyst uses to review/work each case and determine the action that needs to be taken. A Case Management System usually allows the analyst to see recent transactions on the account, and provides the analyst with contact information so an attempt can be made to reach the cardholder. The Case Management System gives the fraud analyst enough information to make a decision in the event that a cardholder cannot be reached. C.Authorization Systems decide whether or not to approve or decline a transaction. The most basic Authorization Systems will make this decision using elements such as: Available Credit Limit/Account Balance Current Status (is the account overdue?) Open/Closed Indicator Expiration Date Check CVV Validation While the Neural Network Model can help predict which accounts/ transactions are most likely to be fraudulent, it requires followup action to verify the activity or to decide whether or not an account should be blocked. The other components of the Fraud Detection System and Authorization System help credit unions and processors take action to identify and stop fraudulent activity as quickly as possible. IV. CONCLUSION Credit card fraud is an act of criminal dishonesty. This article has reviewed recent findings in the credit card field. This paper has identified the different types of fraud, such as bankruptcy fraud, counterfeit fraud, theft fraud, application fraud and behavioral fraud, and discussed measures to detect them. Such measures have included neural networks. References [1]. Aleskerov, E., Freisleben, B. & B Rao CARDWATCH: A Neural Network-Based Database Mining System for Credit Card Fraud Detection, Proc. of the IEEE/IAFE on Computational Intelligence for Financial Engineering, [2]. Anderson, R The Credit Scoring Toolkit: theory and practice for retail credit risk management and decision automation. New York: Oxford University Press. [3]. APACS, Association for Payment Cleaning Services, no date. Card Fraud Facts and Figures Available at: s_and_figures.html (Accessed: December 2007). [4]. Bentley, P., Kim, J., Jung. G. & J Choi Fuzzy Darwinian Detection of Credit Card Fraud, Proc. of 14 th Annual Fall Symposium of the Korean Information Processing Society. [5]. Bolton, R. & Hand, D Statistical Fraud Detection: A Review. Statistical Science, 17; [6]. Brause R., Langsdorf T. & M Hepp. 1999a. Credit card fraud detection by adaptive neural data mining, Internal Report 7/99 (J. W. Goethe-University, Computer Science Department, Frankfurt, Germany). [7]. Brause, R., Langsdorf, T. & M Hepp. 1999b. Neural Data Mining for Credit Card Fraud Detection, Proc. of 11 th IEEE International Conference on Tools with Artificial Intelligence. 63

5 [8]. Caminer, B Credit card Fraud: The Neglected Crime. The Journal of Criminal Law and Criminology, 76; [9]. Chan, P., Fan, W. Prodromidis, A. & S Stolfo Distributed Data Mining in Credit Card Fraud Detection.IEEE Intelligent Systems, 14; [10]. Chan, P., Stolfo, S., Fan, D., Lee, W. & A Prodromidis Credit card fraud detection using meta learning: Issues and initial results, Working notes of AAAI Workshop on AI Approaches to Fraud Detection and Risk Management. [11]. Chepaitis, E Information Ethics Across Information Cultures. Business Ethics: A European Review, 6: 4, [12]. Chiu, C. & Tsai, C A Web Services-Based Collaborative Scheme for Credit Card Fraud Detection. Proc. Of 2004 IEEE International Conference on e-technology, e-commerce and e- Service. [13]. Clarke, M Fraud and the Politics of Morality. Business Ethics: A European Review, 3: 2, [14]. Dorronsoro, J. Ginel, F. Sanchez, C. & C Cruz Neural Fraud Detection in Credit Card Operations. IEEE Transactions on Neural Networks, 8; [15]. Encyclopedia Britannica, no date. Credit Card. Available at: card (Accessed: October 2008). [16]. Bolton, R. & Hand, D Unsupervised Profiling Methods for Fraud Detection, Credit Scoring and Credit Control VII. 64