IP Library Granted Patent US 10,068,088
Granted Patent B2
US 10,068,088 · App. 15/048,021 · Granted Sep 4, 2018

Method, computer program and system that uses behavioral biometric algorithms

Inventors: Neil Costigan (Lulea, SE); Ingo Deutschmann (Merseburg, DE); Tony Libell (Lulea, SE); Johanna Skarpman Munter (Lulea, SE); Peder Nordström (Lulea, SE)
Assignee: BehavioSec
G06F21/55G06F21/316G06F21/32G06K9/00355G06K9/00899G06K9/6296G06F2221/2133
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,068,088
App. No.
15/048,021
Granted
Sep 4, 2018
Kind
B2
Abstract

A computer-implemented method, computer program product, and system for determining whether a user exhibits machine behavior, or does not exhibit human-like behavior, thereby to authenticate the user for access to a software service.

Claims (39)

1. A computer-implemented method comprising:

receiving, via at least one input module of a user-device, an input-set of data, necessary for a request for access to a software service, said input-set of data consciously generated by a user,

gathering from said user-device a data-set of data dependent at least on an unconscious behavior of said user when using said user-device said data-set including randomized elements,

evaluating, based on said data-set, whether said user exhibits human-like behavior or machine-like behavior, said evaluating including detecting a Gaussian distribution of said randomized elements, thereby to identify said data-set as exhibiting machine-like behavior,

when said evaluating determines that said user exhibits machine-like behavior, or does not determine that said user exhibits human-like behavior, executing a bot-routine.

2. The method of claim 1 , wherein said data-set comprises at least one of dwell time data, trajectory data, pressure data, acceleration data, deceleration data, velocity data, data relating to randomness of input, data relating to uniformity of actions, jitter data, pulsation data, distortion data, and life indicator data.

3. The method of claim 1 , wherein said evaluating comprises detecting at least characteristic of said user-device, and using known features of user-devices having said at least one characteristic as part of said evaluating, said at least one characteristic including at least one of a type of computer platform, a type of browser, and an operating system used by said user in said user-device.

4. The method of claim 1 , wherein said data-set comprises trajectory information relating to a trajectory traveled by a pointing element between two keys of the user-device, and said evaluating comprises identifying at least one of specific keys crossed by said trajectory, a lag-time between key touching being greater than a threshold, deviation from a straight line or curve, and jitter present during traversal of said trajectory, thereby to identify said data-set as exhibiting human-like behavior.

5. The method of claim 1 , wherein said data-set comprises information relating to pressure applied by said user to keys of said user-device, and said evaluating comprises detecting at least one of excessive uniformity in said applied pressure and excessive uniformity in the portion of the key to which pressure is applied, thereby to identify said data-set as exhibiting machine-like behavior.

6. The method of claim 1 , wherein said gathering takes place prior to said user providing said input-set of data.

7. The method of claim 1 , wherein said evaluating comprises evaluating whether a human would be physically able to provide said data in said data-set, and, if said human would not be physically able to provide said data, carrying out said executing step.

8. The method of claim 1 , wherein said evaluating comprises actively triggering at least one distortion, expected to cause a specific reaction, discernable in said data-set, if said user-device is operated by said human, thereby to identify whether said user exhibits said human-like behavior.

9. The method of claim 1 , wherein said bot-routine includes at least one of:

interrupting transmission of said input-set from said user-device to said software service;

identifying said user by obtaining additional user information or system information;

storing said obtained additional user information or system information in hashed form in a list of users suspected not to be human;

erasing said input-set from said user-device;

creating a counter tracking the number of times wrong input data was provided for the user, or updating said counter; and

if said counter exceeds a threshold, or increases rapidly, storing said obtained additional user information or system information in hashed form in a black list of users.

10. The method of claim 1 , wherein said evaluating comprises comparing the number of characters in an input field of a data collection form with the number of characters in said data input, thereby to determine at least one of a score and a confidence of the received input data.

11. A computer program product comprising a non-transitory computer-usable medium including instructions which, when executed by a computer, cause the computer to:

receive an input-set of data, necessary for a request for access to a software service, said input-set of data consciously generated by a user,

gather from a user-device operated by said user a data-set of data dependent at least on an unconscious behavior of said user when using said user-device, said data-set including randomized elements,

evaluate, based on said data-set, whether said user exhibits human-like behavior or machine-like behavior, said instructions to evaluate cause the computer to detect a Gaussian distribution of said randomized elements, thereby to identify said data-set as exhibiting machine-like behavior, and

when said evaluation indicates that said user exhibits machine-like behavior, or does not indicate that said user exhibits human-like behavior, execute a bot-routine.

12. The computer program product of claim 11 , wherein said data-set comprises at least one of dwell time data, trajectory data, pressure data, acceleration data, deceleration data, velocity data, data relating to randomness of input, data relating to uniformity of actions, jitter data, pulsation data, distortion data, and life indicator data.

13. The computer program product of claim 11 , wherein said instructions cause the computer to detect at least characteristic of said user-device, and to use known features of user-devices having said at least one characteristic in order to perform said evaluation, said at least one characteristic including at least one of a type of computer platform, a type of browser, and an operating system used by said user in said user-device.

14. The computer program product of claim 11 , wherein said instructions cause said computer to actively trigger at least one distortion, expected to cause a specific reaction, discernable in said data-set, if said user-device is operated by said human, thereby to evaluate whether said user exhibits said human-like behavior.

15. A computer system comprising:

at least one input module adapted to receive input consciously provided by a user;

at least one sensor adapted to gather data from a user, the user not being conscious of said data being gathered; and

one or more processors, the processors being configured to:

receive from said at least one input module an input-set of data, necessary for a request for access to a software service, said input-set of data consciously generated by a user,

gather from said at least one sensor a data-set of data dependent at least on a unconscious behavior of said user when using a user-device, said data-set including randomized elements,

evaluate, based on the data-set, whether said user exhibits a human-like behavior or a machine-like behavior, said evaluation including detecting a Gaussian distribution of said randomized elements, thereby to identify said data-set as exhibiting machine-like behavior, and

when said evaluation indicates that said user exhibits machine-like behavior, or does not indicate that said user exhibits human-like behavior, execute a bot-routine.

16. The computer system of claim 15 , wherein said data-set comprises at least one of dwell time data, trajectory data, pressure data, acceleration data, deceleration data, velocity data, data relating to randomness of input, data relating to uniformity of actions, jitter data, pulsation data, distortion data, and life indicator data.

17. The computer system of claim 15 , wherein said processor is programmed to detect at least characteristic of said user-device, and to use known features of user-devices having said at least one characteristic in order to perform said evaluation, said at least one characteristic including at least one of a type of computer platform, a type of browser, and an operating system used by said user in said user-device.

18. The computer system of claim 15 , wherein said processor is programmed to actively trigger at least one distortion, expected to cause a specific reaction, discernable in said data-set, if said user-device is operated by said human, thereby to evaluate whether said user exhibits said human-like behavior.

Assignments (3)
RELEASE OF SECURITY INTEREST Recorded May 2, 2022
From: SILICON VALLEY BANK
To: BEHAVIOSEC, INC.
Reel/Frame 059778/0709 →
INTELLECTUAL PROPERTY SECURITY AGREEMENT Recorded Mar 1, 2021
From: BEHAVIOSEC INC.
To: SILICON VALLEY BANK
Reel/Frame 055442/0889 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 9, 2016
From: COSTIGAN, NEIL, DR; DEUTSCHMANN, INGO; LIBELL, TONY; MUNTER, JOHANNA SKARPMAN
To: BEHAVIOSEC
Reel/Frame 037932/0035 →
Continuity (3)
Continuation In Part 13866183 · Apr 19, 2013
Provisional Application 61637559 · Apr 24, 2012
Related Publication 20160180083A1 · Jun 23, 2016