Predicting an identity of a person based on an activity history

• US 10,482,386 B2
• Filed: 10/26/2015
• Issued: 11/19/2019
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Intelmate LLC
• Original Assignee: Intelmate LLC
• Inventors: Torgersrud, Richard
• Primary Examiner(s): Hill, Stanley K.
• Application Number:
  US14/923,081
• Publication Number:
  US 20160042293A1
• Time in Patent Office: 1,485 Days
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Systems and methods for predicting an identity of a person are provided. In some aspects, a list of subject activities accessed by a subject person is received. For each of a plurality of stored persons, a stored list of activities accessed with by the stored person is accessed in one or more data repositories. An intersection is calculated between the list of subject activities and the stored list of activities for at least one stored person from the plurality of stored persons. That the subject person is likely to correspond to the at least one stored person from among the plurality of stored persons is predicted, based on the calculated intersection. An indication that the subject person is likely to correspond to the at least one stored person from among the plurality of stored persons is provided.

Claims

1. A computer-implemented method for discovering an identity of a person, the method comprising:

receiving a list of subject activities accessed by a subject person;

accessing one or more data repositories for a plurality of stored lists of activities, each stored list of activities corresponding to a respective stored person;

calculating an intersection between the list of subject activities and a stored list of activities for a stored person, the intersection corresponding to a match between at least one activity on the list of subject activities and at least one activity on the stored list of activities;

predicting, based on the calculated intersection, that the subject person is likely to correspond to the stored person corresponding to the intersected activity;

providing an indication that the subject person is likely to correspond to the stored person;

confirming the subject person corresponds to the stored person using additional identity verification; and

determining a percentage that the predicted correspondence is confirmed.

2. The computer-implemented method of claim 1,

wherein the list of subject activities comprises a contact list from an electronic device of the subject person.

3. The computer-implemented method of claim 1,

wherein the subject person is an inmate in a prison, wherein the one or more data repositories comprise one or more data repositories of prison inmates, and wherein providing the indication that the subject person is likely to correspond to the stored person comprises:

providing, to a client computing device, a photograph of the stored person.

4. The computer-implemented method according to claim 1, wherein the list of subject activities comprises a list of telephone numbers dialed by the subject person, and the stored list of activities comprises a list of telephone numbers dialed by the stored person;

and

wherein predicting that the subject person is likely to correspond to the stored person is based on the stored person having dialed the earliest telephone number dialed by the subject person.

5. The computer-implemented method according to claim 4, further comprising:

excluding telephone numbers of bail bonds offices and telephone numbers of public defenders from the list of subject activities and the stored list of activities from the calculated intersection.

6. The computer-implemented method according to claim 1, wherein the list of subject activities comprises voice samples of the subject person, and the stored list of activities comprises voice samples of the stored person;

and

wherein predicting that the subject person is likely to correspond to the stored person is based on the stored person having a voice pattern matching a voice pattern of the stored person using a voice recognition algorithm.

7. A computer-implemented method for discovering an identity of a person, the method comprising:

receiving a list of subject activities accessed by a subject person;

accessing one or more data repositories for a plurality of stored list of activities, each stored list of activities corresponding to a respective stored person;

calculating an intersection between the list of subject activities and a stored list of activities for a stored person, the intersection corresponding to a match between at least one activity on the list of subject activities and at least one activity on the stored list of activities;

predicting, based on the calculated intersection, that the subject person is likely to correspond to the stored person corresponding to the intersected activity;

providing an indication that the subject person is likely to correspond to the stored person;

wherein the list of subject activities comprises a list of telephone numbers dialed by the subject person, and wherein the stored list of activities comprises a list of telephone numbers dialed by the stored person;

determining, using the list of telephone numbers dialed by the subject person, a threshold number of earliest telephone numbers dialed by the subject person during a first time period;

determining that the threshold number of earliest telephone numbers dialed by the stored person during a second time period is equivalent to the threshold number of earliest telephone number dialed by the subject person during the first time period, wherein predicting that the subject person is likely to correspond to the stored person is based on the threshold number of earliest telephone number dialed by the at least one stored person during the second time period being equivalent to the threshold number of earliest telephone number dialed by the subject person during the first time period.

8. The method of claim 7, wherein the threshold number is 1.

9. The method of claim 7, wherein the threshold number of the earliest telephone numbers dialed by the subject person during a first time period comprises a threshold proportion of the earliest telephone numbers dialed by the subject person during the first time period.

10. The computer-implemented method according to claim 7, further comprising:

confirming the subject person corresponds to the stored person using additional identity verification;

determining the percentage that the predicted correspondence is confirmed; and

adjusting the threshold number of earliest telephone number dialed by the subject person during the first period to maximize the percentage that the predicted correspondence is confirmed.

11. A non-transitory computer-readable medium comprising instructions which, when executed by one or more computing devices, cause the one or more computing devices to implement a method, the method comprising:

receiving a list of subject activities accessed by a subject person;

accessing one or more data repositories for a plurality of stored list of activities, each stored list of activities corresponding to a respective stored person;

calculating an intersection between the list of subject activities and a stored list of activities for a stored person, the intersection corresponding to a match between at least one activity on the list of subject activities and at least one activity on the stored list of activities;

predicting, based on the calculated intersection, that the subject person is likely to correspond to the stored person corresponding to the intersected activity;

providing an indication that the subject person is likely to correspond to the stored person;

confirming the subject person corresponds to the stored person using additional identity verification; and

determining a percentage that the predicted correspondence is confirmed.

12. The non-transitory computer-readable medium of claim 11,

wherein the list of subject activities comprises funding sources depositing money into a financial account of the subject person, and wherein the stored list of activities comprises funding sources depositing money into a financial account of the stored person.

13. The non-transitory computer-readable medium of claim 11,

wherein the list of subject activities comprises a contact list from an electronic device of the subject person.

14. The non-transitory computer-readable medium of claim 11, wherein the list of subject activities comprises a list of telephone numbers dialed by the subject person, and wherein the stored list of activities comprises a list of telephone numbers dialed by the stored person;

and

wherein predicting that the subject person is likely to correspond to the stored person is based on the stored person having dialed the earliest telephone number dialed by the subject person.

15. The non-transitory computer-readable medium of claim 11, wherein the method further comprises:

excluding telephone numbers of bail bonds offices and telephone numbers of public defenders from the list of subject activities and the stored list of activities from the calculated intersection.

16. The non-transitory computer-readable medium of claim 11, wherein the list of subject activities comprises voice samples of the subject person, and the stored list of activities comprises voice samples of the stored person;

and

wherein predicting that the subject person is likely to correspond to the stored person is based on the stored person having a voice pattern matching a voice pattern of the stored person using a voice recognition algorithm.

17. A non-transitory computer-readable medium comprising instructions which, when executed by one or more computing devices, cause the one or more computing devices to implement a method, the method comprising:

receiving a list of subject activities accessed by a subject person;

accessing one or more data repositories for a plurality of stored list of activities, each stored list of activities corresponding to a respective stored person;

calculating an intersection between the list of subject activities and a stored list of activities for a stored person, the intersection corresponding to a match between at least one activity on the list of subject activities and at least one activity on the stored list of activities;

predicting, based on the calculated intersection, that the subject person is likely to correspond to the stored person corresponding to the intersected activity;

providing an indication that the subject person is likely to correspond to the stored person, wherein the list of subject activities comprises a list of telephone numbers dialed by the subject person, and wherein the stored list of activities comprises a list of telephone numbers dialed by the stored person;

determining, using the list of telephone numbers dialed by the subject person, a threshold number of earliest telephone numbers dialed by the subject person during a first time period;

determining that the threshold number of earliest telephone numbers dialed by the stored person during a second time period is equivalent to the threshold number of earliest telephone number dialed by the subject person during the first time period, wherein predicting that the subject person is likely to correspond to the stored person is based on the threshold number of earliest telephone number dialed by the stored person during the second time period being equivalent to the threshold number of earliest telephone numbers dialed by the subject person during the first time period.

18. The non-transitory computer-readable medium of claim 17, wherein the method further comprises:

confirming the subject person corresponds to the stored person using additional identity verification; and

determining the percentage that the predicted correspondence is confirmed.