METHOD AND APPARATUS FOR A MERCHANT PROFILE BUILDER

• US 20070203732A1
• Filed: 02/21/2007
• Published: 08/30/2007
• Est. Priority Date: 02/24/2006
• Status: Active Grant

• Current Assignee: Fair Isaac Corporation
• Original Assignee: Fair Isaac Corporation
• Inventors: Gancarz, Gregory, Zhang, Bo, Milana, Joseph, Yang, Linyu, Li, Danfeng, Zhang, Zuohua, GRIEGEL, David
• Granted Patent: US 8,567,669 B2
• Time in Patent Office: Days
• US Class Current: 705/75
• CPC Class Codes: G06Q 20/401 Transaction verification
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A merchant profile builder (MPB) is provided. The complete merchant profiling process is split into three pieces: Data Preprocessing, Weekly Summary Calculation, and Merchant Profile Calculation. In the Data Preprocessing stage, authorization data and daily fraud data are processed on a weekly basis to yield: an authorization extract file and fraud authorization data, each of which are sorted by a merchant key. In the Weekly Summary Calculation stage, the files output by the Data Preprocessing stage are used as input to compile basic statistics for each merchant based only on data for one week. There are separate weekly summaries for authorizations and fraud authorizations. In the Merchant Profile Calculation stage, the weekly summaries from the 16 most recent weeks are combined together and averages are calculated and output a set of merchant profiles and a set of default profiles.

Claims

1. A computer implemented method of determining a merchant profile data set to be used in a fraud prediction system that generates a score for a transaction in real-time, which score is transmitted to a transaction entity to allow for either completion or termination of the transaction, comprising the steps of:

receiving a payment card authorization data set from a financial institution or from an association of financial institutions containing authorized transactions;

receiving a daily fraud data set containing transactions that are determined to be fraud;

comparing transactions in the payment card authorization data set with those in the daily fraud data set and identifying matching transactions that are in both the payment card authorization data set and the daily fraud data set;

generating an authorization extract file, said authorization extract file sorted by a merchant key;

generating one or more fraud authorization files comprising the identified matching transactions, said one or more fraud authorization files sorted by a merchant key;

compiling statistics for each merchant for each week by applying the generated authorization extract file and using the compiled statistics to generate a weekly summary authorization data set;

compiling statistics for each merchant for each week by applying the generated fraud authorization extract file and using the compiled statistics to generate a weekly summary fraud authorization data set;

combining weekly summaries from the most recent weeks of a predetermined number and calculating averages;

outputting and storing a merchant profile data set based on the weekly summary authorization data and based on the weekly summary fraud authorization data set, said merchant profile data set to be used in a fraud prediction system that generates a score for a transaction in real-time, which score is transmitted to a transaction entity to allow for completion or termination of the transaction; and

outputting and storing a default merchant profile data set based on the calculated averages, said default profile data set to be used in the fraud prediction system that generates the score for a transaction in real-time, which score is transmitted to the transaction entity to allow for completion or termination of the transaction.

2. The method of claim 1, further comprising the step of:

generating merchant profile variables that are intensive variables, independent of the size of the merchant.

3. The method of claim 1, further comprising the step of:

generating fraud rate variables, pick-up rate, and average and standard-deviation transaction amounts.

4. The method of claim 1, further comprising the step of:

generating conditional fraud rates which are rates defined as separate fraud rates for separate characteristics.

5. The method of claim 1, further comprising the step of:

synchronizing the authorizations and fraud authorizations based on their respective transaction dates and ensuring the fraud authorization count in the numerator and the authorization count in the denominator come from the same time period.

6. The method of claim 1, further comprising the step of:

generating merchant profile variables based on merchant country code; and

generating merchant default profile variables based on merchant category code (MCC) and on merchant country code.

7. The method of claim 1, further comprising the step of:

giving equal weight to data from the predetermined number of most recent weeks and removing older weeks for said combining weekly summaries and said calculating averages.

8. The method of claim 1, further comprising the step of:

segmenting the authorizations and fraud authorizations into the following individual segments for which to calculate independent sets of statistics;

cross border, card not present;

cross border, card present;

domestic, card not present;

domestic, card present, chip entry mode;

domestic, card present, swiped entry mode without PIN;

domestic, card present, swiped entry mode with PIN;

domestic, card present, chip fallback entry mode; and

domestic, card present, keyed or unknown entry mode.

9. The method of claim 8, further comprising the step of:

providing adaptive segmentation by which the profile variables for just the top four most popular segments are transmitted in the merchant'"'"'s profile.

10. The method of claim 9, further comprising the step of:

transmitting a segmentation indicator as one of the profile variables that communicates to the model which segments are present and in which slots they reside.

11. The method of claim 6, further comprising the step of:

transmitting the MCC-merchant-country default as part of each merchant profile.

12. The method of claim 1, further comprising the step of:

providing bounds on how many declines are analyzed.

13. A computer system of determining a merchant profile data set to be used in a fraud prediction system that generates a score for a transaction in real-time, which score is transmitted to a transaction entity to allow for either completion or termination of the transaction, comprising:

means for receiving a payment card authorization data set from a financial institution or from an association of financial institutions containing authorized transactions;

means for receiving a daily fraud data set containing transactions that are determined to be fraud;

means for comparing transactions in the payment card authorization data set with those in the daily fraud data set and identifying matching transactions that are in both the payment card authorization data set and the daily fraud data set;

means for generating an authorization extract file, said authorization extract file sorted by a merchant key;

means for generating one or more fraud authorization files comprising the identified matching transactions, said one or more fraud authorization files sorted by a merchant key;

means for compiling statistics for each merchant for each week by applying the generated authorization extract file and using the compiled statistics to generate a weekly summary authorization data set;

means for compiling statistics for each merchant for each week by applying the generated fraud authorization extract file and using the compiled statistics to generate a weekly summary fraud authorization data set;

means for combining weekly summaries from the most recent weeks of a predetermined number and calculating averages;

means for outputting and storing a merchant profile data set based on the weekly summary authorization data and based on the weekly summary fraud authorization data set, said merchant profile data set to be used in a fraud prediction system that generates a score for a transaction in real-time, which score is transmitted to a transaction entity to allow for completion or termination of the transaction; and

means for outputting and storing a default merchant profile data set based on the calculated averages, said default profile data set to be used in the fraud prediction system that generates the score for a transaction in real-time, which score is transmitted to the transaction entity to allow for completion or termination of the transaction.

14. The system of claim 13, further comprising:

means for generating merchant profile variables that are intensive variables, independent of the size of the merchant.

15. The system of claim 13, further comprising:

means for generating fraud rate variables, pick-up rate, and average and standard-deviation transaction amounts.

16. The system of claim 13, further comprising:

means for generating conditional fraud rates which are rates defined separate fraud rates for separate characteristics.

17. The system of claim 13, further comprising:

means for synchronizing the authorizations and fraud authorizations based on their respective transaction dates and ensuring the fraud authorization count in the numerator and the authorization count in the denominator come from the same time period.

18. The system of claim 13, further comprising:

means for generating merchant profile variables based on merchant country code; and

means for generating merchant default profile variables based on merchant category code (MCC) and on merchant country code.

19. The system of claim 13, further comprising:

means for giving equal weight to data from the predetermined number of most recent weeks and removing older weeks for said combining weekly summaries and said calculating averages.

20. The system of claim 13, further comprising:

means for segmenting the authorizations and fraud authorizations into the following individual segments for which to calculate independent sets of statistics;

cross border, card not present;

cross border, card present;

domestic, card not present;

domestic, card present, chip entry mode;

domestic, card present, swiped entry mode without PIN;

domestic, card present, swiped entry mode with PIN;

domestic, card present, chip fallback entry mode; and

domestic, card present, keyed or unknown entry mode.

21. The system of claim 20, further comprising:

means for providing adaptive segmentation by which the profile variables for just the top four most popular segments are transmitted in the merchant'"'"'s profile.

22. The system of claim 21, further comprising:

means for transmitting a segmentation indicator as one of the profile variables that communicates to the model which segments are present and in which slots they reside.

23. The system of claim 18, further comprising:

means for transmitting the MCC-merchant-country default as part of each merchant profile.

24. The system of claim 13, further comprising:

means for providing bounds on how many declines are analyzed.