Long-term short-term cascade modeling for fraud detection

Patent metadata

Field	Value
Publication number	US-10832250-B2
Application number	US-201715683094-A
Country	US
Kind code	B2
Filing date	Aug 22, 2017
Priority date	Aug 22, 2017
Publication date	Nov 10, 2020
Grant date	Nov 10, 2020

Abstract

Embodiments disclosed herein are related to computing systems and methods for determining a risk score for a plurality of data transactions. In the embodiments, a first risk score module may receive data transactions. The first risk score module may then determine a first risk score for each of the data transactions. A second risk score module that is different from the first risk score module may receive each of the first risk scores determined by the first risk score module as an input. The second risk score module may determine a second risk score based in part on the input first risk scores for each of the data transactions. The second risk scores may specify if each of the data transactions is to be approved or rejected by the computing system.

First claim

What is claimed is:

1. A computing system for determining a risk score for a plurality of data transactions, the computing system comprising:
at least one processor;
a computer readable hardware storage device having stored thereon computer-executable instructions which, when executed by the at least one processor, cause the computing system to perform the following:
an act of receiving a plurality of data transactions at a first risk score module, the first risk score module being a long-term risk score module with respect to a second risk score module;
an act of the first risk score module determining a first risk score for each of the plurality of data transactions;
an act of the second risk score module that is different from the first risk score module, the second risk score module being a short-term risk score module with respect to the first risk score module, receiving each of the first risk scores determined by the first risk score module as an input; and
an act of the second risk score module determining a second risk score based on the input first risk scores and one or more other risk parameters for each of the plurality of data transactions; and
based on the second risk score, performing, in real time, at least one of approving, rejecting, or causing to be further reviewed each of the data transactions, by a decision module of the computing system.

2. The computing system according to claim 1 , wherein the first risk score module is trained using data that is collected for a first period of time.

3. The computing system according to claim 2 , wherein the first period of time is at least one year.

4. The computing system according to claim 2 , wherein the training data that is collected comprises one or more of chargeback data from an outside evaluator or review data from an outside review module.

5. The computing system according to claim 1 , wherein the second risk score module is trained using data that is collected for a second period of time that is shorter than a first period of time that data that is used to train the first module is collected.

6. The computing system according to claim 5 , wherein the second period of time is one month or less.

7. The computing system according to claim 5 , wherein the training data that is collected comprises one or more of chargeback data from an outside evaluator or review data from an outside review module.

8. The computing system according to claim 1 , further comprising:
an act of the second risk module receiving one or more risk parameters that are related to each of the plurality of data transactions and that are indicative of a risk level of each of the plurality of data transactions; and
an act of determining the second risk scores based in part on the input first risk scores for each of the plurality of data transactions and on the received one or more risk parameters scores.

9. The computing system according to claim 1 , further comprising:
an act of calibrating the second risk scores based at least in part on the input first risk.

10. A method for determining a risk score for a plurality of data transactions, the method comprising:
a computing system comprising at least one processor performing an act of receiving a plurality of data transactions at a first risk score module;
the at least one processor performing an act of the first risk score module determining a first risk score for each of the plurality of data transactions, wherein the first risk core module is a long-term risk score module with respect to a second risk score module;
the at least one processor performing an act of the second risk score module that is different from the first risk score module receiving each of the first risk scores determined by the first risk score module as an input, wherein the second risk score module is a short-term risk score module with respect to the first risk score module; and
the at least one processor performing an act of the second risk score module determining a second risk score based on the input first risk scores and one or more other risk parameters for each of the plurality of data transactions; and
based on the second risk score, the at least one processor performing, in real time, at least one of approving, rejecting, or causing to be further reviewed each of the data transactions, by a decision module of the computing system.

11. The method according to claim 10 , wherein the first risk score module is trained using data that is collected for a first period of time.

12. The method according to claim 11 , wherein the first period of time is at least one year.

13. The method according to claim 10 , wherein the second risk score module is trained using data that is collected for a second period of time that is shorter than a first period of time that data that is used to train the first module is collected.

14. The method according to claim 13 , wherein the second period of time is one month or less.

15. The method according to claim 13 , wherein the training data that is collected comprises one or more of chargeback data from an outside evaluator or review data from an outside review module.

16. The method according to claim 10 , further comprising:
an act of the second risk module receiving one or more risk parameters that are related to each of the plurality of data transactions and that are indicative of a risk level of each of the plurality of data transactions; and
an act of determining the second risk scores based in part on the input first risk scores for each of the plurality of data transactions and on the received one or more risk parameters scores.

17. The method according to claim 10 , further comprising:
an act of calibrating the second risk scores based at least in part on the input first risk.

18. A computing system for determining a risk score for a plurality of data transactions, the computing system comprising:
at least one processor;
a computer readable hardware storage device having stored thereon computer-executable instructions which, when executed by the at least one processor, cause the computing system to instantiate the following:
a first risk score module, wherein the first risk score module is a long-term risk score module with respect to a second risk score module, that is configured to:
receive a plurality of data transactions; and
determine a first risk score for each of the plurality of data transactions; and
the second risk score module that is different from the first risk score module, wherein in the second risk score module is a short-term risk score module with respect to the first risk score module, and that is configured to:
receive each of the first risk scores determined by the first risk score module as an input; and
determine a second risk score based on the input first risk scores and one or more other risk parameters for each of the plurality of data transactions; and
a decision module that is configured to, based on the second risk score, in real time, at least one of approve, reject, or cause to be further reviewed each of the data transactions.

19. The computing system according to claim 18 , wherein the computer readable hardware storage device has stored thereon computer-executable instructions which, when executed by the at least one processor, cause the computing system to instantiate the following:
a first training module that is configured to:
collect first data over a first period of time, the first data based in part on the first risk scores determined by the first risk score

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Jia Yuting
• Wang Shoou-Jiun
• Marcjan Cezary
• Nanduri Jayaram N M

Classifications

• G06Q40/02 Primary

  Banking, e.g. interest calculation or account maintenance (credit or loans G06Q40/03) · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q20/4016 Primary

  involving fraud or risk level assessment in transaction processing · CPC title