METHOD FOR GENERATING A HUMAN LIKENESS SCORE

• US 20120246737A1
• Filed: 03/02/2012
• Published: 09/27/2012
• Est. Priority Date: 03/24/2011
• Status: Active Grant

• Current Assignee: Imperva Incorporated (Thales SA)
• Original Assignee: Ayah LLC (Thales SA)
• Inventors: Paxton, Tyler James, Tatoris, Reid Michael, Blackmer, Benjamin Crispell, Aleksa, Adam, VandenBrink, Stuart
• Granted Patent: US 8,776,173 B2
• Time in Patent Office: Days
• US Class Current: 726/27
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 21/31 User authentication
• CPC Class Codes: G06F 21/316 by observing the pattern of computer usage, e.g. typical user behaviour
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 21/36 by graphic or iconic representation
• CPC Class Codes: G06F 2221/2103 Challenge-response
• CPC Class Codes: G06F 2221/2133 Verifying human interaction, e.g., Captcha
• CPC Class Codes: H04L 63/083 using passwords cryptographic mechanisms or cryptographic arrangements for entity authentication using a predetermined code H04L9/3226

Abstract

One embodiment of the invention is a method utilizing a CAPTCHA to generate a human likeness score including blocks: a) receiving a user solution to the CAPTCHA; b) receiving a user interaction pattern descriptive of an interaction undertaken by the user, through a graphical interface of the CAPTCHA, to achieve the user solution; c) determining the accuracy of the user solution; d) comparing the user interaction pattern against an interaction model generated from interaction patterns of previous users; e) calculating the human likeness score based upon the determination of block c) and the comparison of block d), wherein the human likeness score lies within a continuum of human likeness scores.

Claims

1. A method utilizing a CAPTCHA to generate a human likeness score, the CAPTCHA provided through a graphical interface, the method comprising blocks:

a) receiving a user solution to the CAPTCHA;

b) receiving a user interaction pattern descriptive of an interaction undertaken by the user, through the graphical interface, to achieve the user solution;

c) determining the accuracy of the user solution;

d) comparing the user interaction pattern against an interaction model generated from interaction patterns of previous users;

e) calculating the human likeness score based upon the determination in block c) and the comparison in block d), wherein the human likeness score lies within a continuum of human likeness scores.

2. The method of claim 1, further comprising comparing the human likeness score to a threshold likeness score to verify that the user is human.

3. The method of claim 2, further comprising adding the user interaction pattern to a database of user interaction patterns, including interaction patterns of previous users.

4. The method of claim 2, including determining the user to be human if the human likeness score is greater than the threshold likeness score.

5. The method of claim 2, including determining the user to be automated software if the human likeness score is less than the threshold likeness score.

6. The method of claim 2, further comprising transmitting, to a publisher, the result of the comparison of the human likeness score to the threshold likeness score.

7. The method of claim 2, further comprising adjusting the threshold likeness score based upon an input from a publisher.

8. The method of claim 1, further comprising providing a server accessible by a publisher to receive the human likeness score.

9. The method of claim 1, further comprising transmitting the human likeness score to a publisher.

10. The method of claim 1, further comprising calculating an automated software likeness score based upon the determination in block c) and the comparison in block d), wherein the automated software likeness score lies within a continuum of automated software likeness scores.

11. The method of claim 1, wherein the interaction undertaken by the user, through the graphical interface, to achieve the user solution is selected from the group consisting of:

a cursor movement;

a mouse click;

a keystroke;

a voice command;

a joystick movement;

a touch;

a facial expression; and

a gesture.

12. The method of claim 1, further comprising receiving user data that is distinct from the CAPTCHA, wherein block e) includes calculating a human likeness score that is further based upon the user data.

13. The method of claim 12, wherein the user data includes at least one of:

a CAPTCHA previously attempted by the user;

an Internet Protocol address of the user; and

a cookie associated with the user.

14. The method of claim 1, wherein block e) includes implementing a decision tree to calculate the human likeness score.

15. The method of claim 1, wherein:

block c) includes generating a first confidence group based upon the user solution;

block d) includes generating a second confidence group based upon the user interaction pattern; and

block e) includes determining the reliability of the first and the second confidence groups and compiling the first and the second confidence groups, based upon the determined reliability thereof, to generate the human likeness score.

16. The method of claim 15, further comprising:

accessing user data, including at least one of a CAPTCHA previously attempted by the user, an Internet Protocol address of the user, and a cookie associated with the user; and

generating a third confidence group based upon the user data;

wherein block e) further includes determining the reliability of the third confidence group and compiling the first, the second, and the third confidence groups, based upon the determined reliability thereof, to generate the human likeness score.

17. The method of claim 1, wherein the continuum of human likeness scores is a continuum of number values, and wherein block e) includes selecting a number value, from the continuum of number values, that correlates with a calculated confidence that the user is human.

18. The method of claim 1, further comprising generating the interaction model by aggregating interaction patterns of previous users, wherein the interaction model is indicative of a characteristic of an interaction undertaken by a human through the graphical interface.

19. The method of claim 1, wherein the interaction model is generated from interaction patterns of previous users, the previous users including at least one known human.

20. The method of claim 1, further comprising isolating the user interaction pattern from the user interaction;

wherein the user interaction is a path-type input provided through the graphical interface, the user interaction pattern is the noise in a portion of the input path, and block d) includes comparing the noise of the user interaction pattern with the interaction model that defines noise typical of a human interaction.

21. The method of claim 1, further comprising isolating the user interaction pattern from the user interaction;

wherein the user interaction is a path-type input provided through the graphical interface, the user interaction pattern is a direction change in a portion of the input path, and block d) includes comparing the direction change of the user interaction pattern with the interaction model that defines a direction change typical of a human interaction.

22. The method of claim 1, further comprising isolating the user interaction pattern from the user interaction;

wherein the user interaction is a path-type input provided through the graphical interface, the user interaction pattern is an input velocity of a portion of the input path, and block d) includes comparing the input velocity of the user interaction pattern with the interaction model that defines an input velocity typical of a human interaction.

23. The method of claim 1, further comprising isolating the user interaction pattern from the user interaction;

wherein the user interaction pattern is a characteristic of a response to a moving object within the graphical interface, and wherein block d) includes comparing the response of the user interaction pattern with the interaction model that defines a characteristic of a response typical of a human interaction.

24. The method of claim 23, wherein the characteristic of the response is selected from the group consisting of:

a lag in direction change of an input relative to a direction change of the moving object;

an input path leading up to selection of the moving object;

an input path following release of the moving object; and

input timing following a change in distance between the object and a second object depicted within the graphical interface.

25. The method of claim 1, further comprising generating the CAPTCHA, including:

defining the graphical interface and an object depicted therein;

assigning an instruction for completing the CAPTCHA; and

pseudorandomly assigning a characteristic of the object depicted within the graphical interface to differentiate the CAPTCHA from a previously-generated CAPTCHA.

26. The method of claim 25, including pseudorandomly assigning at least one of an initial location, an initial orientation, an initial speed, and an initial direction of motion to the object.

27. The method of claim 25, wherein generating the CAPTCHA follows receiving a request, from a publisher, for the CAPTCHA.

28. The method of claim 27, wherein the request defines a generic format for the CAPTCHA.

29. The method of claim 27, further comprising receiving an asset from the publisher, wherein the asset is incorporated into the CAPTCHA as an advertisement.

30. The method of claim 25, wherein the instruction is related to a characteristic of the object, including at least one of the shape, the color, the size, the type, and the context of the object.

31. The method of claim 25, wherein the interaction undertaken by the user includes manipulating the object within the graphical interface.

32. The method of claim 31, wherein block b) includes tracking a manipulation of the object within the graphical interface by the user.

33. The method of claim 32, including capturing both a timing characteristic and a path characteristic of the manipulation of the object within the graphical interface by the user.

34. The method of claim 25, further comprising transmitting a portion of the CAPTCHA to a publisher.

35. The method of claim 34, wherein generating the CAPTCHA includes configuring the CAPTCHA to be presented through an interface selected from the group consisting of:

a website;

a mobile application;

a web-based forum; and

a web-based email client.

36. The method of claim 1, further comprising adding the user solution to a catalog of user preferences, wherein the CAPTCHA defines a survey and the user solution is indicative of a user preference.

37. The method of claim 36, wherein adding the user solution to the catalog of user preferences is only performed if the human likeness score is greater than a threshold likeness score.

38. The method of claim 1, wherein the user interaction pattern comprises a keystroke, wherein the state of a key and the timing of the keystroke are compared against the interaction model.

39. A method utilizing a CAPTCHA to verify that a user is human, the CAPTCHA provided through a graphical interface, the method comprising blocks:

a) receiving a user solution to the CAPTCHA;

b) receiving a user interaction pattern descriptive of an interaction undertaken by the user, through the graphical interface, to achieve the user solution;

c) receiving user data that is distinct from the CAPTCHA and the user interaction pattern;

d) determining the accuracy of the user solution;

e) comparing the user interaction pattern against an interaction model generated from interaction patterns of previous users;

f) analyzing the user data for automated software indicators;

g) calculating a human likeness score based upon the determination in block d), the comparison in block e), and the analysis in block f), wherein the human likeness score lies within a continuum of human likeness scores; and

h) comparing the human likeness score to a threshold likeness score to verify that the user is human.

40. The method of claim 39, wherein the user data includes at least one of:

a CAPTCHA previously attempted by the user;

an Internet Protocol address of the user; and

a cookie associated with the user.

41. The method of claim 39, further comprising determining the reliability of the user solution, the user interaction pattern, and the user data to indicate that the user is human.