APPARATUS AND METHOD FOR STATISICAL USER AUTHENTICATION USING INCREMENTAL USER BEHAVIOR

• US 20120185916A1
• Filed: 06/28/2011
• Published: 07/19/2012
• Est. Priority Date: 01/14/2011
• Status: Abandoned Application

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: CHAE, Seung-Chul, LEE, Sun-Jae
• Application Number:
  US13/170,818
• Publication Number:
  US 20120185916A1
• Time in Patent Office: Days
• US Class Current: 726/2
• CPC Class Codes: G06F 21/316 by observing the pattern of computer usage, e.g. typical user behaviour

Abstract

Provided are an apparatus and method for statistical user identification that improves a user'"'"'s convenience while ensuring security. The apparatus may store a confidence value that statistically represents whether a user is identified as a user that has permission to use a terminal, based on a user event that occurs when the user manipulates the terminal. The apparatus may determine whether to execute a user requested application by comparing the confidence value of the user with a reference value that is defined for the requested application.

Claims

1. An apparatus for statistical user identification, the apparatus comprising:

an update unit configured to update a confidence value that statistically represents whether a user is identified as being a user that has permission to use a terminal, based on a user event that occurs when the user manipulates the terminal; and

a statistical authentication unit configured to, in response to a user requesting execution of an application that requests authentication, determine whether to execute the requested application by comparing the confidence value with a reference value that is defined for the requested application.

2. The apparatus of claim 1, further comprising a storage unit configured to store the confidence value, a user model that is generated based on feature information about a manipulation pattern that is obtained when the user manipulates the terminal, and a reference value defined for each application.

3. The apparatus of claim 2, wherein the update unit updates the confidence value by increasing or decreasing the confidence value based on the comparison between the user event and the user model.

4. The apparatus of claim 2, wherein the update unit decreases the confidence value if a user event is not detected within a predetermined period of time.

5. The apparatus of claim 2, wherein the update unit applies feature information of a user event that occurs after the updated confidence value exceeds a predetermined threshold value, to the user model.

6. The apparatus of claim 1, wherein the statistical authentication unit accepts the execution of the application, if the confidence value exceeds the reference value.

7. The apparatus of claim 1, wherein the statistical authentication unit denies the execution of the application, if the confidence value is below the reference value.

8. The apparatus of claim 1, wherein the statistical authentication unit demands additional authentication from the user, if the confidence value is below the reference value.

9. The apparatus of claim 8, wherein the statistical authentication unit calculates a difference between the confidence value and the reference value, if the confidence value is below the reference value, and demands additional authentications from the user based on the calculated difference.

10. A method of statistical user identification, the method comprising:

updating a confidence value that statistically represents whether a user is identified as being a user that has permission to use a terminal, based on a user event that occurs when the user manipulates the terminal; and

upon occurrence of a user event requesting execution of an application that requests authentication, determining whether to execute the requested application by comparing the confidence value with a reference value that is defined for the requested application.

11. The method of claim 10, wherein the updating comprises updating the confidence value by increasing or decreasing the confidence value based on a comparison between the user event and the user model.

12. The method of claim 10, wherein the updating comprises decreasing the confidence value, if a user event is not detected within a predetermined period of time.

13. The method of claim 10, further comprising applying feature information of a user event that occurs after the updated confidence value exceeds a predetermined threshold value, to the user model.

14. The method of claim 10, wherein the determining of execution of the application comprises executing the application, if the confidence value exceeds the reference value.

15. The method of claim 10, wherein the determining of execution of the application comprises not executing the application, if the confidence value is below the reference value.

16. The method of claim 10, wherein the determining of execution of the application comprises requesting an additional authentication of the user, if the confidence value is below the reference value.

17. A method for statistical user identification, the method comprising:

detecting a user event that occurs when a user manipulates a terminal;

comparing feature information of the detected user event with feature event of a user model and updating a confidence value that statistically represents whether a user has permission to use the terminal, based on a comparison result of the feature information;

determining whether the detected user event is a user event requesting execution of an application which is executable in the terminal and which requests authentication; and

determining whether to execute the application by comparing the updated confidence value with a reference value that is a previously defined for each application, if the detected user event is the user event requesting execution of the application.

18. The method of claim 17, further comprising:

generating an initial user model based on feature information about a manipulation pattern that is obtained when the user manipulates the terminal; and

generating an initial confidence value that statistically represents whether a user has permission to use the terminal.

19. A terminal for statistical identification of a user that manipulates the terminal, the terminal comprising:

a storage unit configured to store a confidence value that statistically represents whether the user has permission to use a respective application; and

a statistical authentication unit configured to determine whether the user has permission to execute a requested application by comparing the user'"'"'s confidence value with a reference value that is defined for the requested application.

20. The terminal of claim 19, wherein the user'"'"'s confidence value is N, a first application requested by the user has a reference value that is greater than N, and the statistical authentication unit determines to deny execution of the application based on a comparison of the user'"'"'s confidence value and the reference value.

21. The terminal of claim 20, wherein a second application requested by the user has a reference value that is less than N, and the statistical authentication unit determines to execute the application based on a comparison of the user'"'"'s confidence value and the reference value.

22. The terminal of claim 19, wherein the confidence value is a number between 0% and 100%, 0% represents the minimum confidence value the terminal can have for the user, and 100% represents the maximum confidence value that the terminal can have for the user.

23. The terminal of claim 19, further comprising:

one or more sensors configured to sense user manipulation of the terminal; and

an update unit configured to update the confidence value of the user based on the sensed user'"'"'s manipulation of the terminal.

24. The terminal of claim 23, wherein the one or more sensors sense at least one of pressure applied by the user to a touch screen of the terminal and speed at which the user inputs characters on the touch screen of the terminal, and the update unit updates the confidence value based on the sensed data.