Method for subject classification using a pattern recognition input device

Patent metadata

Field	Value
Publication number	US-9329699-B2
Application number	US-201113279279-A
Country	US
Kind code	B2
Filing date	Oct 22, 2011
Priority date	Oct 22, 2010
Publication date	May 3, 2016
Grant date	May 3, 2016

Abstract

The present invention provides a device and method for classifying a user using pattern recognition of an input device. A series of the keystroke objects are received via the user input interface. A typing signature is determined for the series of keystroke objects using the processor by analyzing the key attributes of the series of keystroke objects using a pattern recognition algorithm. The typing signature is compared to one or more user typing signatures stored in the memory using the processor. The user is classified based on whether or not the typing signature is statistically similar to one of the stored typing signatures.

First claim

What is claimed is:

1. A method for classifying a user using pattern recognition of an input device comprising the steps of:
providing an input device having a user input interface, one or more sensors, a processor and a memory, wherein (a) a keystroke object is created whenever a single user input is received by the user input interface, and (b) the keystroke object includes three or more key attributes comprising (1) a vector of variable length representing pressure data measured by the one or more sensors during entry of the single user input, (2) timing data measured by the one or more sensors during entry of the single user input, and (3) a unique symbol, function or command associated with the single user input;
receiving a series of the keystroke objects via the user input interface;
determining a typing signature for the series of keystroke objects using the processor by analyzing the three or more key attributes for each keystroke object within the series of keystroke objects using a pattern recognition algorithm;
comparing the typing signature to one or more user typing signatures stored in the memory using the processor, wherein each stored user typing signature includes a user specific inconsistency/consistency factor for each of the key attributes of the series of keystroke objects that either permits or penalizes a variability in the each of the key attributes of the stored typing signature; and
classifying the user based on whether or not the typing signature is statistically similar to one of the stored user typing signature, and updating the stored user typing signature for the user whenever the typing signature is statistically similar to the stored user typing signature of the user.

2. The method of claim 1 , further comprising the step of storing the user typing signature for a particular user and a particular series of keystroke objects via the input device.

3. The method of claim 2 , further comprising the steps of:
receiving the particular series of keystroke objects via the user input interface; and
determining the user typing signature for the particular series of keystroke objects using the processor by analyzing the key attributes of the particular series of keystroke objects using the pattern recognition algorithm.

4. The method of claim 1 , further comprising the step of providing the classification of the user to the user.

5. The method of claim 1 , wherein:
the user input interface comprises a keyboard, a keypad, one or more keys, one or more buttons, a touch screen or a touch pad;
the series of keystroke objects comprise a user name, a user identifier, a password, a passcode, a passphrase, a response to a security question or a combination thereof;
the input device is communicably coupled or integrated into a computer, a laptop, a workstation, a phone, a user device, an electronic tablet, an entertainment device, a security device, a machine, a vehicle or a combination thereof; and
the classification comprises an authentication or verification of the user, a classification of the user as a member of a group or a combination thereof.

6. The method as recited in claim 1 , wherein the receiving, determining and comparing steps are repeated for two or more series of keystroke objects before performing the classifying step.

7. The method of claim 1 , wherein:
the timing data comprises a hold time for the single user input and a flight time between the single user input and a previous single user input, if any; and
the pressure data comprises a maximum pressure and one or more other pressures measured during the hold time for the single user input.

8. The method of claim 1 , wherein the pressure data comprises a maximum pressure, two or more other pressures measured during a hold time for the single user input, an average pressure measured during the hold time for the single user input, a continuous stream or function describing pressure measured during the hold time for the single user input or a combination thereof.

9. The method of claim 1 , wherein the series of keystroke objects is preset, continuous, periodic or random.

10. The method of claim 1 , further comprising the step of fitting a polynomial curve to the keystroke object for use by the pattern recognition algorithm.

11. The method of claim 1 , further comprising the step of calculating a mean and a standard deviation of each keystroke object for use by the pattern recognition algorithm.

12. The method of claim 1 , wherein the stored user typing signature comprises an adaptive signature that encompasses a mood and a posture of the user and the input device to gradually become more specific over time.

13. The method of claim 1 , wherein a variable threshold value can be adjusted to change a true positive rate and/or a true negative rate of the classification.

14. The method of claim 1 , wherein a threshold for the statistical similarity determination changes over time, changes based on a location of the input device, changes based on one or more security parameters or a combination thereof.

15. The method as recited in claim 1 , wherein the one or more stored user typing signatures further comprise one or more stored past false entries, and the comparison step further comprises penalizing the typing signature that is statistically similar to one of the stored past false entries.

16. The method as recited in claim 1 , wherein the comparison and classification steps allows a deviation in the unique symbol, function or command of the keystroke objects in the typing signature as compared to the unique symbol, function or command of the keystroke objects in the stored user typing signature.

17. The method as recited in claim 1 , further comprising the step of, after the user has been authenticated or validated, continuously validating the user throughout a user session when one or more sets of keystroke objects corresponding to one or more predefined keywords are entered by the user.

18. A pattern recognition input device comprising
a user input interface;
one or more sensors coupled to the user input interface that create a keystroke object whenever a single user input is received by the user input interface, wherein the keystroke object includes three or more key attributes comprising (1) a vector of variable length representing pressure data measured by the one or more sensors during entry of the single user input, (2) timing data measured by the one or more sensors during entry of the single user input, and (3) a unique symbol, function or command associated with the single user input;
a processor communicably coupled to the one or more sensors;
a memory communicably coupled to the processor; and
wherein the processor (a) receives a series of the keystroke objects, (b) determines a typing signature for the series of keystroke objects by analyzing the three or more key attributes for each keystroke object within [[of]] the series of keystroke objects using a pattern recognition algorithm, (c) compares the typing signature to one or more user typing signatures stored in the memory, wherein each stored user typing signature includes a user specific inconsistency/consistency factor for each of the key attributes of the series of keystroke objects that either permits or penalizes a variability in each one of the key attributes of the stored typing signature and (d) classifies the user based on whether or not the typing signature is statistically similar to one of the stored user typing signatures, and updates the stored user typing signature for the user whenever the

Assignees

• Allen Jeffrey David
• Howard John Joseph
• Thornton Mitchell Aaron
• Univ Southern Methodist

Inventors

• Allen Jeffrey David
• Howard John Joseph
• Thornton Mitchell Aaron

Classifications

• G06F3/023 Primary

  Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes · CPC title

• G06F21/32

  using biometric data, e.g. fingerprints, iris scans or voiceprints · CPC title

• G06F21/316

  by observing the pattern of computer usage, e.g. typical user behaviour · CPC title