Model-based prediction of an optimal convenience metric for authorizing transactions

• US 10,332,113 B2
• Filed: 11/18/2015
• Issued: 06/25/2019
• Est. Priority Date: 11/19/2014
• Status: Active Grant

• Current Assignee: EyeLock LLC (VOXX International Corporation)
• Original Assignee: EyeLock LLC (VOXX International Corporation)
• Inventors: Hanna, Keith, Teverovskiy, Mikhail, Aggarwal, Manoj, Makthal, Sarvesh
• Primary Examiner(s): Shrestha, Bijendra K
• Application Number:
  US14/945,023
• Publication Number:
  US 20160140567A1
• Time in Patent Office: 1,315 Days
• CPC Class Codes: G06Q 20/20 Point-of-sale [POS] network systems
• CPC Class Codes: G06Q 20/32 using wireless devices
• CPC Class Codes: G06Q 20/40145 Biometric identity checks
• CPC Class Codes: G06Q 20/4016 involving fraud or risk level assessment in transaction processing

Abstract

The present disclosure describes systems and methods for authorization. The method may include accessing, by an authorization engine for a transaction by a user, an activity pattern model of the user from a database. The activity pattern model of the user may be indicative of a geospatial behavior of the user over time. The authorization engine may determine a set of sensors available for facilitating the transaction, each of the sensors assigned with a usability value prior to the transaction. The authorization engine may access an activity pattern model of the sensors, the activity pattern model of the sensors indicative of geospatial characteristics of one or more of the sensors over time. The authorization engine may determine a convenience metric for each of a plurality of subsets of the sensors, using the activity pattern model of the user, the activity pattern model of the sensors, and usability values of corresponding sensors. Each of the plurality of subsets may include corresponding sensors that can be used in combination to facilitate the transaction. The authorization engine may select, using the determined convenience metrics, a subset from the plurality of subsets to use for the transaction.

Claims

1. A method of authorization, the method comprising:

accessing, by an authorization engine comprising one or more processors, for a transaction by a user, an activity pattern model of the user from a database, the activity pattern model of the user indicative of a geospatial behavior of the user over time;

determining, by the authorization engine, a set of sensors available for facilitating the transaction, each of the sensors assigned with a usability value prior to the transaction;

accessing, by the authorization engine, an activity pattern model of the sensors, the activity pattern model of the sensors indicative of geospatial characteristics of one or more of the sensors over time;

determining, by the authorization engine, a convenience metric for each of a plurality of subsets of the sensors, using the activity pattern model of the user, the activity pattern model of the sensors, and usability values of corresponding sensors, each of the plurality of subsets comprising corresponding sensors that can be used in combination for the transaction; and

selecting, by the authorization engine using the determined convenience metrics, a subset from the plurality of subsets for the transaction.

2. The method of claim 1, wherein the usability value of a corresponding sensor is indicative of an amount of user effort to provide input to the corresponding sensor.

3. The method of claim 1, wherein the selected subset of sensors includes a biometric sensor.

4. The method of claim 1, wherein at least one of the activity pattern model of the user or the activity pattern model of the sensors comprises a component that is dynamically updated during part of the transaction.

5. The method of claim 1, wherein at least one of the activity pattern model of the user or the activity pattern model of the tensors comprises at least one of:

a sequence model, a velocity model, a periodic model, a habitat model, or a neural network model.

6. The method of claim 1, wherein determining the convenience metric for each of the plurality of subsets further comprises determining the convenience metric for each of the plurality of subsets using outputs of at least one of the sensors obtained prior to the transaction.

7. The method of claim 1, wherein determining the convenience metric for each of the plurality of subsets further comprises determining the convenience metric for each of the plurality of subsets using a device-specific metric of at least one of the sensors, the device-specific metric comprising a false accept rate, a false reject rate or quality of acquired biometrics.

8. The method of claim 1, wherein determining the convenience metric for each of the plurality of subsets further comprises determining the convenience metric for each of the plurality of subsets using outputs of at least one sensor different from sensors of the first set, obtained prior to the transaction.

9. The method of claim 1, wherein selecting the subset from the plurality of subsets to use for the transaction comprises:

determining, by the authorization engine for each of the plurality of subsets using a corresponding convenience metric, an expected income from authorizing the transaction; and

selecting, by the authorization engine, a subset from the plurality of subsets to use for the transaction, the selected subset determined with an expected income that is higher than that of others of the plurality of subsets of sensors.

10. The method of claim 9, wherein the expected income incorporates at least one of:

a benefit or income from a correct determination to authorize the transaction, a potential loss due to an incorrect determination to authorize the transaction, or an expected loss of income from losing the user as a customer.

11. A system of authorization, the system comprising:

a set of sensors available for facilitating a transaction by a user, each of the sensors comprising an input sensing hardware interface;

a database configured to store at least one of an activity pattern model of the user and an activity pattern model of the sensors; and

an authorization engine comprising one or more processors, configured to;

access, for a transaction, the activity pattern model of the user from the database, the activity pattern model of the user indicative of a geospatial behavior of the user over time;

determine the set of sensors available for facilitating the transaction, each of the sensors assigned with a usability value prior to the transaction;

access the activity pattern model of the sensors, the activity pattern model of the sensors indicative of geospatial characteristics of one or more of the sensors over time;

determine a convenience metric for each of a plurality of subsets of the sensors, using the activity pattern model of the user, the activity pattern model of the sensors, and usability values of corresponding sensors, each of the plurality of subsets comprising corresponding sensors that can be used in combination for the transaction; and

select, using the determined convenience metrics, a subset from the plurality of subsets for the transaction.

12. The system of claim 11, wherein the usability value of a corresponding sensor is indicative of an amount of user effort to provide input to the corresponding sensor.

13. The system of claim 11, wherein the selected subset of sensors includes a biometric sensor.

14. The system of claim 11, wherein at least one of the activity pattern model of the user or the activity pattern model of the sensors comprises a component that is dynamically updated during part of the transaction.

15. The system of claim 11, wherein at least one of the activity pattern model of the user or the activity pattern model of the sensors comprises at least one of:

a sequence model, a velocity model, a periodic model, a habitat model, or a neural network model.

16. The system of claim 11, wherein the authorization engine is configured to determine the convenience metric for each of the plurality of subsets using outputs of at least one of the sensors obtained prior to the transaction.

17. The system of claim 11, wherein the authorization engine is configured to determine the convenience metric for each of the plurality of subsets using a device-specific metric of at least one of the sensors, the device-specific metric comprising a false accept rate, a false reject rate or quality of acquired biometrics.

18. The system of claim 11, wherein the authorization engine is configured to determine the convenience metric for each of the plurality of subsets using outputs of at least one sensor different from sensors of the first set, obtained prior to the transaction.

19. The system of claim 11, wherein the authorization engine is configured to select the subset from the plurality of subsets to use for the transaction, comprising:

determining, by the authorization engine for each of the plurality of subsets using a corresponding convenience metric, an expected income from authorizing the transaction; and

selecting, by the authorization engine, a subset from the plurality of subsets to use for the transaction, the selected subset determined with an expected income that is higher than that of others of the plurality of subsets of sensors.

20. The system of claim 19, wherein the expected income incorporates at least one of:

a benefit or income from a correct determination to authorize the transaction, a potential loss due to an incorrect determination to authorize the transaction, or an expected loss of income from losing the user as a customer.