Method and apparatus for evaluating fraud risk in an electronic commerce transaction

• US 8,019,678 B2
• Filed: 11/03/2010
• Issued: 09/13/2011
• Est. Priority Date: 05/30/2001
• Status: Expired due to Fees

• Current Assignee: Cybersource Corporation (Visa, Inc.)
• Original Assignee: Cybersource Corporation (Visa, Inc.)
• Inventors: Hu, Hung-Tzaw, Wright, William
• Primary Examiner(s): Kramer; James A
• Assistant Examiner(s): Gottschalk; Martin A
• Application Number:
  US12/939,080
• Publication Number:
  US 20110047044A1
• Time in Patent Office: 314 Days
• Field of Search: 705/35, 705/38, 705/39
• US Class Current: 705/38
• CPC Class Codes: G06Q 20/04 Payment circuits
• CPC Class Codes: G06Q 20/10 specially adapted for electronic funds transfer [EFT] systems; specially adapted for home banking systems
• 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 20/403 Solvency checks
• CPC Class Codes: G06Q 30/0609 Buyer or seller confidence or verification
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G07F 7/08 by coded identity card or credit card or other personal identification means without personal verification means G07F7/02

Abstract

Transaction information is received and applied to multiple fraud risk mathematical models that each produce a respective raw score, which are transformed with respective sigmoidal transform functions to produce optimized likelihood of fraud risk estimates to provide to a merchant. Respective fraud risk estimates are combined using fusion proportions associated with the respective risk estimates, producing a single point risk estimate, which is transformed with a sigmoidal function to produce an optimized single point risk estimate. The sigmoidal functions approximate a relationship between risk estimates produced by fraud risk detection models and a percentage of transactions associated with respective risk estimates; the relationship is represented in terms of real-world distributions of fraudulent and non-fraudulent transaction.

Claims

1. A method for evaluating fraud risk in an electronic commerce transaction and providing a representation of the fraud risk to a merchant using electronic communication, the method comprising:

one or more processors generating and storing two or more fraud risk mathematical models, each model having a corresponding distribution of fraudulent transactions and a corresponding distribution of non-fraudulent transactions;

defining for each mathematical model a first point corresponding to a risk estimate at which a percentage of fraudulent transactions begins to have a non-zero value;

a second point corresponding to a risk estimate at which a percentage of non-fraudulent transactions begins to have a zero count; and

a third point corresponding to a risk estimate at which the percentage of fraudulent transactions equals the percentage of non-fraudulent transactions;

the one or more processors receiving information about a transaction and performing;

for each fraud risk mathematical model of the two or more fraud risk mathematical models, applying the information about the transaction to said each fraud risk mathematical model, said each fraud risk mathematical model producing a corresponding raw score;

transforming the corresponding raw score into a corresponding risk estimate for said each fraud risk mathematical model;

blending the corresponding risk estimate of said each fraud risk mathematical model into a single fraud score for the transaction including;

for said each risk estimate corresponding to said each fraud risk mathematical model, determining a corresponding fraud risk zone, wherein the fraud risk zones are determined based on the first point, the second point, and the third point defined by a particular fraud risk mathematical model;

determining a weighting factor to apply to said each risk estimate based on the corresponding fraud risk zone;

modifying said each risk estimate using that weighting factor.

2. The method of claim 1, wherein the scoring authority is a first scoring authority and further comprising:

in response to the first risk estimate corresponding to the particular fraud risk mathematical model falling within a range of values between risk estimates corresponding to the third point and the second point respectively that are defined by the particular fraud risk mathematical model, establishing one of the fraud risk mathematical models as a second scoring authority;

wherein the first scoring authority is a different fraud risk mathematical model than the first scoring authority.

3. The method of claim 2 further comprising

transforming the single fraud score for the transaction to produce an optimized single risk estimate for the transaction.

4. The method of claim 2, wherein at least one corresponding risk estimate is based on a corresponding statistical model from a set of one or more statistical mathematical models and at least one other corresponding risk estimate is based on a corresponding heuristic mathematical model;

wherein the at least one heuristic model produces a corresponding risk estimate based on a weighted score of answers to a set of questions with discrete responses.

5. The method of claim 4, wherein the particular fraud risk mathematical model is a heuristic model.

6. The method of claim 4, wherein the particular fraud risk mathematical model is a statistical model.

7. The method of claim 4, wherein the first scoring authority is a heuristic model and the second scoring authority is a statistical model.

8. The method of claim 4, wherein each weight applied to a score for each test result of the set of questions is based on merchant interest in said each test result.

9. A non-transitory computer-readable volatile or non-volatile medium storing one or more sequences of instructions for evaluating fraud risk in an electronic commerce transaction and providing a representation of the fraud risk to a merchant using electronic communication, which instructions, when executed by one or more processors, cause the one or more processors to perform:

generating and storing two or more fraud risk mathematical models, each model having a corresponding distribution of fraudulent transactions and a corresponding distribution of non-fraudulent transactions;

defining for each mathematical model a first point corresponding to a risk estimate at which a percentage of fraudulent transactions begins to have a non-zero value;

a second point corresponding to a risk estimate at which a percentage of non-fraudulent transactions begins to have a zero count; and

a third point corresponding to a risk estimate at which the percentage of fraudulent transactions equals the percentage of non-fraudulent transactions;

receiving information about a transaction and performing;

for each fraud risk mathematical model of the two or more fraud risk mathematical models, applying the information about the transaction to said each fraud risk mathematical model, said each fraud risk mathematical model producing a corresponding raw score;

transforming the corresponding raw score into a corresponding risk estimate for said each fraud risk mathematical model;

blending the corresponding risk estimate of said each fraud risk mathematical model into a single fraud score for the transaction including;

for said each risk estimate corresponding to said each fraud risk mathematical model, determining a corresponding fraud risk zone, wherein the fraud risk zones are determined based on the first point, the second point, and the third point defined by a particular fraud risk mathematical model;

determining a weighting factor to apply to said each risk estimate based on the corresponding fraud risk zone;

modifying said each risk estimate using that weighting factor.

10. The non-transitory computer-readable volatile or non-volatile medium of claim 9, wherein the scoring authority is a first scoring authority and further comprising instructions, which instructions, when executed by one or more processors, cause the one or more processors to perform:

in response to the first risk estimate corresponding to the particular fraud risk mathematical model falling within a range of values between risk estimates corresponding to the third point and the second point respectively that are defined by the particular fraud risk mathematical model, establishing one of the fraud risk mathematical models as a second scoring authority;

wherein the first scoring authority is a different fraud risk mathematical model than the first scoring authority.

11. The non-transitory computer-readable volatile or non-volatile medium of claim 10 further comprising instructions, which instructions, when executed by one or more processors, cause the one or more processors to perform transforming the single fraud score for the transaction to produce an optimized single risk estimate for the transaction.

12. The non-transitory computer-readable volatile or non-volatile medium of claim 10, wherein at least one corresponding risk estimate is based on a corresponding statistical model from a set of one or more statistical mathematical models and at least one other corresponding risk estimate is based on a corresponding heuristic mathematical model;

wherein the at least one heuristic model produces a corresponding risk estimate based on a weighted score of answers to a set of questions with discrete responses.

13. The non-transitory computer-readable volatile or non-volatile medium of claim 12, wherein the particular fraud risk mathematical model is a heuristic model.

14. The non-transitory computer-readable volatile or non-volatile medium of claim 12, wherein the particular fraud risk mathematical model is a statistical model.

15. The non-transitory computer-readable volatile or non-volatile medium of claim 12, wherein the first scoring authority is a heuristic model and the second scoring authority is a statistical model.

16. The non-transitory computer-readable volatile or non-volatile medium of claim 12, wherein each weight applied to a score for each test result of the set of questions is based on merchant interest in said each test result.