Device security via swipe pattern recognition

Patent metadata

Field	Value
Publication number	US-10185817-B2
Application number	US-201615184732-A
Country	US
Kind code	B2
Filing date	Jun 16, 2016
Priority date	Jun 16, 2016
Publication date	Jan 22, 2019
Grant date	Jan 22, 2019

Abstract

One embodiment provides a method for ensuring device security via swipe pattern recognition, the method comprising: utilizing at least one processor to execute computer code that performs the steps of: receiving, using a touch device, at least one swipe input of a user; determining, using the at least one processor, if the at least one swipe input matches a known swipe pattern of the user, the match requiring exceeding a match confidence level; and responsive to said determining, executing an action associated with the touch device. Other aspects are described and claimed.

First claim

What is claimed is:

1. A method for ensuring device security via swipe pattern recognition, the method comprising:
utilizing at least one processor to execute computer code that performs the steps of:
receiving, using a touch device, at least one swipe input of a user during interaction with the touch device, wherein the at least one swipe input comprises a plurality of swipes;
determining, using the at least one processor, if the at least one swipe input matches a known swipe pattern of the user, the match requiring exceeding a match confidence level, wherein the known swipe pattern is determined via a machine learning process that monitors touch input of a user during interaction with the touch device and groups the touch input into swipe pattern types based upon characteristics of the touch input, the swipe pattern types corresponding to a use case of the touch input;
the determining comprising comparing the at least one swipe input to the swipe pattern type corresponding to the use case of the at least one swipe input and determining a similarity between the at least one swipe input and the swipe pattern type, wherein the similarity is based on a determined sequence of the plurality of swipes within a predetermined window of time and wherein the similarity is based on at least one of: swipe speed, swipe acceleration, swipe length, swipe shape, and swipe pressure; and
responsive to said determining, executing an action associated with the touch device.

2. The method of claim 1 ,
wherein the machine learning comprises grouping historical swipes, received during normal use of the touch device, based on at least one similar characteristic.

3. The method of claim 1 , wherein the action comprises securing the touch device against unauthorized use.

4. The method of claim 1 , comprising generating at least one user profile corresponding to one or more subsets of the previously received swipes.

5. The method of claim 4 , wherein the action comprises: prompting, using the processor, a user to enter identification credentials.

6. The method of claim 1 , comprising:
detecting, using a sensor, a location of the touch device;
wherein the action is responsive to the detected location.

7. The method of claim 1 , comprising:
detecting, using a sensor, a movement of the touch device;
wherein the action is responsive to the detected movement.

8. An apparatus for ensuring device security via swipe pattern recognition, the apparatus comprising:
a touch surface;
at least one processor; and
a computer readable storage medium having computer readable program code embodied therewith and executable by the at least one processor, the computer readable program code comprising:
computer readable program code that receives, at the touch surface, at least one swipe input during interaction with the touch device, wherein the at least one swipe input comprises a plurality of swipes;
computer readable program code that determines, using the at least one processor, if the at least one swipe input matches a known swipe pattern, the match requiring exceeding a match confidence level, wherein the known swipe pattern is determined via a machine learning process that monitors touch input of a user during interaction with the touch device and groups the touch input into swipe pattern types based upon characteristics of the touch input, the swipe pattern types corresponding to a use case of the touch input;
the determining comprising comparing the at least one swipe input to the swipe pattern type corresponding to the use case of the at least one swipe input and determining a similarity between the at least one swipe input and the swipe pattern type, wherein the similarity is based on a determined sequence of the plurality of swipes within a predetermined window of time and wherein the similarity is based on at least one of: swipe speed, swipe acceleration, swipe length, swipe shape, and swipe pressure; and
computer readable program code that, responsive to said determination, executes an action associated with the apparatus.

9. A computer program product for ensuring device security via swipe pattern recognition, the computer program product comprising:
a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising:
computer readable program code that receives, at the touch surface, at least one swipe input during interaction with the touch device, wherein the at least one swipe input comprises a plurality of swipes;
computer readable program code that determines, using the at least one processor, if the at least one swipe input matches a known swipe pattern, the match requiring exceeding a match confidence level, wherein the known swipe pattern is determined via a machine learning process that monitors touch input of a user during interaction with the touch device and groups the touch input into swipe pattern types based upon characteristics of the touch input, the swipe pattern types corresponding to a use case of the touch input;
the determining comprising comparing the at least one swipe input to the swipe pattern type corresponding to the use case of the at least one swipe input and determining a similarity between the at least one swipe input and the swipe pattern type, wherein the similarity is based on a determined sequence of the plurality of swipes within a predetermined window of time and wherein the similarity is based on at least one of: swipe speed, swipe acceleration, swipe length, swipe shape, and swipe pressure; and
computer readable program code that responsive to said determination, executes an action associated with the touch device.

10. The computer program product of claim 9 ,
wherein the machine learning comprises grouping historical swipes, received during normal use of the touch device, based on at least one similar characteristic.

11. The computer program product of claim 9 , wherein the action comprises securing the touch device against unauthorized use.

12. The computer program product of claim 9 , comprising generating at least one user profile corresponding to one or more subsets of the previously received swipes.

13. The computer program product of claim 12 , wherein the action comprises: prompting, using the processor, a user to enter identification credentials.

14. The computer program product of claim 9 , comprising:
detecting, using a sensor, a location of the touch device;
wherein the action is responsive to the detected location.

15. The computer program product of claim 9 , comprising:
detecting, using a sensor, a movement of the touch device;
wherein the action is responsive to the detected movement.

16. A method of passive privacy preserving authentication for touch devices, the method comprising:
receiving, at a touch device, a sequence of strokes during interaction with the touch device, wherein the sequence of strokes comprises a plurality of strokes;
grouping, based on stoke characteristics, similar strokes;
generating, based on the grouped similar strokes, a user alphabet;
receiving, at the touch surface, at least one additional stroke;
determining, using a processor, if the at least one additional stroke shares one or more characteristics with the user alphabet, wherein the user alphabet is determined via a machine learning process that monitors touch input of a user during interaction with the touch device and groups the touch input into swipe pattern types based upon characteristics of the touch input, the swipe pattern types corresponding to a use case of the sequence of

Assignees

• IBM

Inventors

• Gupta Megha
• Kumar Apurva
• Kundu Ashish
• Nanavati Amit Anil

Classifications

• G06N99/005

  Physics · mapped topic

• G06F21/36 Primary

  by graphic or iconic representation · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F3/03547

  Touch pads, in which fingers can move on a surface · CPC title

• G06F3/04883

  for inputting data by handwriting, e.g. gesture or text · CPC title