FRAUD REDUCTION SYSTEM FOR TRANSACTIONS

• US 20110196791A1
• Filed: 10/13/2010
• Published: 08/11/2011
• Est. Priority Date: 02/08/2010
• Status: Active Grant

• Current Assignee: Visa International Service Association (Visa, Inc.)
• Original Assignee: Visa International Service Association (Visa, Inc.)
• Inventors: Dominguez, Benedicto Hernandez
• Granted Patent: US 10,089,683 B2
• Time in Patent Office: Days
• US Class Current: 705/44
• CPC Class Codes: G06Q 20/40 Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
• CPC Class Codes: G06Q 20/4016 involving fraud or risk level assessment in transaction processing
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

A system, apparatus, and method for reducing fraud in payment or other transactions by providing issuers with a warning that a transaction being processed for authorization is potentially fraudulent. In some embodiments, the present invention processes data obtained from a consumer authentication process that is used in card not present (CNP) transactions to determine characteristics or indicia of fraud from previous transactions. The characteristics or indicia of fraud can be used to generate a set of fraud detection rules or another form of fraud assessment model. A proposed transaction can then be evaluated for potential fraud using the fraud assessment model.

Claims

1. An apparatus for processing a transaction, comprising:

a processor programmed to execute a set of instructions;

a memory coupled to the processor for storing the set of instructions; and

the set of instructions stored in the memory, wherein when executed by the programmed processor the apparatus processes the transaction by receiving consumer authentication data for card not present transactions from an authentication history server;

receiving transaction data for a known fraudulent transaction;

processing the authentication data and transaction data to generate a rule operative to identify whether a proposed transaction is fraudulent;

receiving transaction data for the proposed transaction;

applying the generated rule to the transaction data for the proposed transaction; and

based on the application of the generated rule, determining if the proposed transaction is fraudulent.

2. The apparatus of claim 1, further comprising instructions that when executed by the programmed processor generate a message identifying the proposed transaction as fraudulent.

3. The apparatus of claim 2, further comprising instructions that when executed by the programmed processor provide the generated message to a Directory Server.

4. The apparatus of claim 2, further comprising instructions that when executed by the programmed processor provide the generated message to an Access Control Server.

5. The apparatus of claim 1, wherein the authentication data stored in the authentication history server includes a record of a consumer authentication request for a consumer'"'"'s account that occurred as part of a card not present transaction that was attempted by the consumer using the account.

6. The apparatus of claim 1, wherein the instructions which when executed by the programmed processor process the authentication data and transaction data to generate a rule include instructions which when executed by the programmed processor generate the rule by applying a linear regression model.

7. The apparatus of claim 1, wherein the message identifying the proposed transaction as fraudulent is generated prior to the generation of an authorization request message for the proposed transaction.

8. A system for reducing fraud in transactions, comprising:

an authentication history server storing data records of consumer authentication operations for card not present transactions; and

a risk management engine operative to

process the data records and process transaction data for known fraudulent transactions to determine one or more rules for identifying a fraudulent transaction;

receive transaction data for a proposed transaction;

receive data from the authentication history server for an account corresponding to the proposed transaction;

apply the one or more rules for identifying a fraudulent transaction to the data received from the authentication history server and the transaction data for the proposed transaction; and

generate a message identifying the proposed transaction as fraudulent based on the outcome of applying the one or more rules.

9. The system of claim 8, wherein the risk management engine is further operative to provide the generated message to a Directory Server.

10. The system of claim 8, wherein the risk management engine is further operative to provide the generated message to an Access Control Server.

11. The system of claim 8, wherein the message identifying the proposed payment transaction as fraudulent is generated prior to the generation of an authorization request message for the proposed transaction.

12. The system of claim 8, wherein the one or more rules for identifying a fraudulent transaction are determined by applying a linear regression model to the data records and the transaction data for known fraudulent transactions.

13. A method of reducing fraud in transactions, comprising:

receiving consumer authentication data for card not present transactions from an authentication history server;

receiving transaction data for a known fraudulent transaction; and

processing the authentication data and transaction data in an electronic data processor to generate a rule operative to identify whether a proposed transaction is fraudulent.

14. The method of claim 13, further comprising:

receiving transaction data for the proposed transaction;

receiving data from the authentication history server for an account corresponding to the proposed transaction;

applying the rule to the received transaction data and the data from the authentication history server;

determining if the proposed transaction is fraudulent based on the application of the rule; and

generating a message identifying the proposed transaction as fraudulent if it is determined that the proposed transaction is fraudulent.

15. The method of claim 14, further comprising providing the generated message to a Directory Server.

16. The method of claim 14, further comprising providing the generated message to an Access Control Server.

17. The method of claim 14, wherein the message identifying the proposed transaction as fraudulent is generated prior to the generation of an authorization request message for the proposed transaction.

18. The method of claim 13, wherein processing the authentication data and transaction data in an electronic data processor to generate the rule further comprises applying a linear regression model to the authentication data and transaction data.

19. The method of claim 14, wherein determining if the proposed transaction is fraudulent based on the application of the rule further comprises determining if the proposed transaction is fraudulent based on application of the rule and by considering the results of one or more of a velocity check, a limit check or a geo-location check.