Systems and methods for identifying biometric information as trusted and authenticating persons using trusted biometric information

• US 9,143,506 B2
• Filed: 09/30/2013
• Issued: 09/22/2015
• Est. Priority Date: 02/13/2013
• Status: Active Grant

• Current Assignee: Daniel Duncan
• Original Assignee: Daniel Duncan
• Inventors: Duncan, Daniel
• Primary Examiner(s): HOFFMAN, BRANDON S
• Application Number:
  US14/040,970
• Publication Number:
  US 20140230032A1
• Time in Patent Office: 722 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: H04L 63/0861 using biometrical features, e.g. fingerprint, retina-scan cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231

Abstract

Systems and methods of identifying biometric information as trusted and authenticating persons using trusted biometric information are provided. Biometric information can be identified as trusted by comparing it to other biometric information and/or based on the age of the biometric information and/or receiving correct answers to questions based on personal information. Once biometric information has been identified as trusted it can be used as part of an authentication process by receiving biometric information from a person to be authenticated and comparing the received biometric information with the trusted biometric information.

Claims

1. A method, comprising:

comparing, by a centralized computer, first and second historical biometric information for a first person;

determining, by the centralized computer, a first similarity measure based on the comparison of the first and second historical biometric information;

identifying, by the centralized computer, the first and second historical biometric information as trusted first and second biometric information when the first similarity measure is greater than or equal to a first similarity measure threshold;

attempting, by the centralized computer, to authenticate a person;

selecting, by the centralized computer, information from the group comprising transaction data, the stored first and second biometric information, and non-biometric for the person to be authenticated;

calculating, by the centralized computer using a predictive model, a high-risk transaction probability using the selected information; and

comparing, by the centralized computer, the calculated high-risk transaction probability to a risk threshold, wherein when the high-risk transaction probability is greater than the risk threshold the request to authenticate is further evaluated to identify potential fraud.

2. The method of claim 1, wherein the selecting, calculating, and comparing steps are performed when the attempt to authenticate has failed, and the further evaluation to identify potential fraud comprises:

posing questions to the person requesting authentication; and

identifying the request to authenticate as fraudulent when answers to the posed questions do not match stored answers to the questions.

3. The method of claim 2, wherein when the questions have been previously posed to a person requesting authentication as the first person, an order of the posed questions or content of the posed questions is different between the two requests for authentication.

4. The method of claim 2, wherein the attempt to authenticate includes the centralized computer receiving third biometric information, and when the attempt to authenticate is identified as a fraudulent attempt the third biometric information is added to a blacklist.

5. The method of claim 4, wherein the first, second, and third biometric information is voice biometric information.

6. The method of claim 4, wherein the third biometric information is near-real-time biometric information.

7. The method of claim 1, wherein the first and second historical biometric information are received from different ones of a plurality of independent sources.

8. The method of claim 7, further comprising:

receiving, by the centralized computer, additional, non-biometric personal information for the first person from one of the plurality of independent sources; and

comparing, by the centralized computer, the additional, non-biometric personal information for the first person with other non-biometric personal information for the first person, wherein the other non-biometric personal information is obtained from a source other than the plurality of independent sources,

wherein the first person is identified as trusted when

the first similarity measure is greater than or equal to the first similarity measure threshold; and

the additional, non-biometric personal information matches the other non-biometric personal information.

9. The method of claim 8, wherein when the additional, non-biometric personal information does not match the other non-biometric personal information, the method further comprises:

informing an information source that provided the third biometric information or an information user that the third biometric information is not associated with the first person.

10. The method of claim 7, further comprising:

receiving, by the centralized computer, an indication whether the first person is in good standing with at least one of the plurality of independent sources for a predetermined period of time,

wherein the first and second biometric information is identified as trusted first and second biometric information when

the first similarity measure is greater than or equal to the first similarity measure threshold; and

the first person is in good standing with at least one of the first or second one of the plurality of independent sources for the predetermined period of time.

11. The method of claim 7, wherein the first and second historical biometric information for the first person is received from the first one and second one of the plurality of independent sources in bulk with biometric information for a plurality of other persons.

12. The method of claim 11, wherein the biometric information for the first person and the plurality of other persons includes historical voice biometric information obtained independent of a process of identifying the person and the plurality of other persons as trusted.

13. The method of claim 1, wherein the selecting, calculating, and comparing steps are performed in response to a determination that the attempt to authenticate the person is in connection with a high risk transaction.

14. The method of claim 13, the method further comprising:

initiating, by the centralized computer, contact with the person to be authenticated for the high risk transaction;

capturing, by the centralized computer, a third biometric information from a person answering the contact; and

authenticating, by the centralized computer, the person answering the contact as the person to be authenticated when the third biometric information has a certain measure of similarity to the first trusted biometric information, the second trusted biometric information, or a composite of at least the first and second trusted biometric information.

15. The method of claim 1, wherein the risk threshold is adjustable based on a level of risk of a transaction for which the person to be authenticated is requesting authentication.

16. The method of claim 1, wherein the first and second historical biometric information are received from the same source and the identification of the first and second historical biometric information as trusted first and second biometric information further accounts for an age of the first and second historical biometric information.

17. The method of claim 1, wherein the non-biometric information for the person to be authenticated includes real-time location information or device characterizing information for the device from which the attempt to authenticate is initiated.

18. The method of claim 1, wherein the attempt to authenticate the person is an authentication in connection with an account, wherein the first person and at least one other person are authorized users of the account.

19. A method, comprising:

receiving, by a centralized computer, personal information for a first person from a plurality of independent sources, wherein the personal information for the first person includes first historical biometric information from a first one of the plurality of independent sources;

receiving, by the centralized computer, a request to authenticate a person, wherein the request to authenticate includes second biometric information;

comparing, by the centralized computer, the second biometric information with the first historical biometric information;

determining, by the centralized computer, a similarity measure based on the comparison of the second biometric information with the first historical biometric information;

selecting, by the centralized computer, information from the group comprising transaction data, the stored first historical biometric information, the received second biometric information, the determined similarity measure, and non-biometric for the person to be authenticated;

calculating, by the centralized computer using a predictive model, a high-risk transaction probability using the selected information; and

comparing, by the centralized computer, the calculated high-risk transaction probability to a risk threshold, wherein when the high-risk transaction probability is greater than the risk threshold the request to authenticate is further evaluated to identify potential fraud.

20. The method of claim 19, wherein the selecting, calculating, and comparing steps are performed when the attempt to authenticate has failed, and the further evaluation to identify potential fraud comprises:

posing questions to the person requesting authentication; and

identifying the request to authenticate as fraudulent when answers to the posed questions do not match stored answers to the questions.

21. The method of claim 20, wherein when the request to authenticate is identified as fraudulent the second biometric information is added to a blacklist.

22. The method of claim 19, wherein the first and second biometric information is voice biometric information.

23. The method of claim 19, wherein the first biometric information for the first person is received from the first one of the plurality of independent sources in bulk with biometric information for a plurality of other persons.

24. The method of claim 19, wherein the second biometric information is near-real-time biometric information.

25. The method of claim 19, wherein the selecting, calculating, and comparing steps are performed in response to a determination that the attempt to authenticate the person is in connection with a high risk transaction.

26. The method of claim 25, the method further comprising:

initiating, by the centralized computer, contact with the person to be authenticated for the high risk transaction;

capturing, by the centralized computer, a third biometric information from a person answering the contact; and

authenticating, by the centralized computer, the person answering the contact as the person to be authenticated when the second and third biometric information each has a certain measure of similarity to the first historical biometric information.

27. The method of claim 19, wherein second biometric information is received by the centralized computer via an interactive voice response (IVR) system and when the similarity measure determined based on the comparison of the second biometric information with the first historical biometric information is less than a similarity measure threshold, the method further comprising:

transferring the person to be authenticated from the IVR system to a live agent;

receiving, during an interaction between the live agent and the person to be authenticated, third biometric information; and

comparing the third biometric information with the first historical biometric information and the second biometric information to identify whether there are any anomalies between the compared biometric information.

28. The method of claim 19, wherein the risk threshold is adjustable based on a level of risk of a transaction for which the person to be authenticated is requesting authentication.

29. The method of claim 19, wherein the selected information includes at least the first historical biometric information and the received second biometric information, and wherein an amount of time between when the first historical biometric was captured and when the received second biometric information was captured is used as a factor in the calculation of the high-risk transaction probability.