User authentication using a mobile device
Machine-learning based user authentication using a mobile device (e.g., using a computerized tool) is enabled. For example, a non-transitory machine-readable medium can comprise executable instructions that, when executed by a processor, facilitate performance of operations, comprising: determining an input received via a mobile device, determining, based on the input and using an authentication model, whether the input threshold matches an input pattern associated with an authorized user profile authorized to access a feature of the mobile device, wherein the input pattern has been determined based on machine learning applied to past inputs at the mobile device other than the input, and wherein the authentication model has been generated based on the machine learning applied to the input pattern, and based on a determination that the input at the mobile device is associated with an authorized user profile, granting access to the feature of the mobile device.
1. A system, comprising:
a processor; and
a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising:
based on an output of a sensor of a mobile device, determining motion data representative of motion of the mobile device, wherein the motion data are generated based on repeated user movements or activities for using the mobile device and repeated to exceed a predetermined threshold count during a preset time window to be recognized as a motion pattern associated with an authorized user profile, and wherein the motion data further comprise a location of the mobile device detected with the repeated user movements or activities, and a time of the repeated user movements or activities, and the motion pattern represents user habits of using the mobile device that becomes a user signature associated with a user of the authorized user profile, and wherein the user signature is not based on biometric information of the user;
determining, based on the motion data and using an authentication model, whether the motion of the mobile device threshold matches the motion pattern associated with the authorized user profile authorized to access a feature of the mobile device, wherein the motion pattern has been determined based on machine learning applied to past motion of the mobile device other than the motion of the mobile device, and wherein the authentication model has been generated based on machine learning applied to the motion pattern;
based on a determination that the motion of the mobile device does not threshold match the motion pattern, blocking access to the feature of the mobile device;
determining an input received at the mobile device, wherein the input received at the mobile device comprises an application accessed for a threshold amount of time during a defined time window, activities correlated to the application over time, reactions correlated to the application over time, or a combination thereof, and wherein the application runs on the mobile device;
determining, based on the input and using the authentication model, whether the input threshold matches an input pattern associated with the authorized user profile, wherein the input pattern has been determined based on the machine learning applied to past inputs at the mobile device other than the input, and wherein the authentication model has been further generated based on the machine learning applied to the input pattern; and
based on a determination that the input does not threshold match the input pattern, blocking the access to the feature of the mobile device.
2. The system of claim 1 , wherein the sensor comprises an accelerometer, and wherein the motion of the mobile device comprises a speed, angle, or motion range of the mobile device.
3. The system of claim 1 , wherein the sensor comprises a pressure sensor, and wherein the motion of the mobile device comprises a degree of force of applied to a touch screen of the mobile device.
4. The system of claim 1 , wherein the operations further comprise:
in response to blocking access to the feature of the mobile device,
generating a prompt for an alternate authentication feature associated with the authorized user profile, and
displaying the prompt via a graphical user interface of the mobile device, wherein the alternate authentication feature comprises a comparison of an input at the mobile device with a defined input known to be associated with the authorized user profile; and
in response to the alternate authentication feature being determined to be completed via the mobile device, unblocking access to the feature of the mobile device.
5. The system of claim 4 , wherein the defined input comprises a prerecorded video clip associated with the authorized user profile, wherein the alternate authentication feature comprises a comparison of the prerecorded video clip and a live stream captured by a camera of the mobile device, and wherein the operations further comprise:
in response to the live stream and the prerecorded video clip being determined to comprise a threshold similarity according to a similarity criterion, unblocking access to the feature of the mobile device.
6. The system of claim 4 , wherein the defined input comprises a prerecorded audio clip associated with the authorized user profile, wherein the alternate authentication feature comprises a comparison of the prerecorded audio clip and a live stream captured by a microphone of the mobile device, and wherein the operations further comprise:
in response to the live stream and the prerecorded audio clip being determined to comprise a threshold similarity according to a similarity criterion, unblocking access to the feature of the mobile device.
7. The system of claim 1 , wherein the feature comprises an application of the mobile device or a hardware component of the mobile device.
8. The system of claim 1 , wherein the activities correlated to the application over time further comprise activities inside visited websites over time, pressure and length of touch associated with the application, a movement of the mobile device associated with the application, or a combination thereof.
9. The system of claim 8 , wherein the input pattern comprises a habitual user input associated with the authorized user profile, and wherein the habitual user input comprises a sequence of inputs received via the mobile device.
10. The system of claim 8 , wherein the input pattern comprises an application accessed for a threshold amount of time during a defined time window.
11. A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
based on an output of a sensor of a mobile device, determining motion data representative of motion of the mobile device, wherein the motion data are generated based on repeated user movements or activities for using the mobile device and repeated to exceed a predetermined threshold count during a preset time window to be recognized as a motion pattern associated with an authorized user profile, and wherein the motion data further comprise a location of the mobile device detected with the repeated user movements or activities, and a time of the repeated user movements or activities, and the motion pattern represents user habits of using the mobile device that becomes a user signature associated with a user of the authorized user profile, and wherein the user signature is not based on biometric information of the user;
determining, based on the motion data and using an authentication model, whether the motion of the mobile device threshold matches the motion pattern associated with the authorized user profile authorized to access a feature of the mobile device, wherein the motion pattern has been determined based on machine learning applied to past motion of the mobile device other than the motion of the mobile device, and wherein the authentication model has been generated based on machine learning applied to the motion pattern;
based on a determination that the motion of the mobile device does not threshold match the motion pattern, blocking access to the feature of the mobile device;
determining an input received via the mobile device, wherein the input received at the mobile device comprises an application accessed for a threshold amount of time during a defined time window, activities correlated to the application over time, reactions correlated to the application over time, or a combination thereof, and wherein the application runs on the mobile device;
determining, based on the input and using the authentication model, whether the input threshold matches an input pattern associated with the authorized user profile authorized to access a feature of the mobile device, wherein the input pattern has been determined based on machine learning applied to past inputs at the mobile device other than the input, and wherein the authentication model has been generated based on the machine learning applied to the input pattern; and
based on a determination that the input at the mobile device is associated with the authorized user profile, granting access to the feature of the mobile device.
12. The non-transitory machine-readable medium of claim 11 , wherein the past inputs at the mobile device are determined using a tracking cookie installed on the mobile device.
13. The non-transitory machine-readable medium of claim 11 , wherein the feature comprises an unlock function, executable by the mobile device and configured to unlock a door of a vehicle communicatively coupled to the mobile device.
14. The non-transitory machine-readable medium of claim 11 , wherein the feature comprises a package release request function, executable by the mobile device and configured to generate a package release request signal and to send the package release request signal to a device associated with a delivery entity, and wherein the package release request function is registered with the delivery entity.
15. The non-transitory machine-readable medium of claim 11 , wherein the feature comprises graphic representation, rendered via a graphical user interface of the mobile device, of vaccine data representative of a vaccine associated with the authorized user profile.
16. The non-transitory machine-readable medium of claim 11 , wherein the feature comprises graphic representation, rendered via a graphical user interface of the mobile device, of an identification card associated with the authorized user profile.
17. A method, comprising:
determining, by a processing system including a processor, a motion data representative of motion of a mobile device based on an output of a sensor of the mobile device, wherein the motion data are generated based on repeated user movements or activities for using the mobile device and repeated to exceed a predetermined threshold count during a preset time window to be recognized as a motion pattern associated with an authorized user profile, and wherein the motion data further comprise a location of the mobile device detected with the repeated user movements or activities, and a time of the repeated user movements or activities, and the motion pattern represents user habits of using the mobile device that becomes a user signature associated with a user of the authorized user profile, and wherein the user signature is not based on biometric information of the user;
determining, based on the motion data and using an authentication model, whether the motion of the mobile device threshold matches the motion pattern associated with the authorized user profile authorized to access a feature of the mobile device, wherein the motion pattern has been determined based on machine learning applied to past motion of the mobile device other than the motion of the mobile device, and wherein the authentication model has been generated based on machine learning applied to the motion pattern;
based on a determination that the motion of the mobile device threshold matches the motion pattern, granting access to the feature of the mobile device;
determining an input received at the mobile device, wherein the input received at the mobile device comprises an application accessed for a threshold amount of time during a defined time window, activities correlated to the application over time, reactions correlated to the application over time, or a combination thereof, and wherein the application runs on the mobile device;
determining, based on the input and using the authentication model, whether the input threshold matches an input pattern associated with the authorized user profile, wherein the input pattern has been determined based on the machine learning applied to past inputs at the mobile device other than the input, and wherein the authentication model has been further generated based on the machine learning applied to the input pattern; and
based on a determination that the input does not threshold match the input pattern, blocking the access to the feature of the mobile device.
18. The method of claim 17 , further comprising:
based on a determination that the motion of the mobile device does not threshold match the motion pattern, denying access to the feature of the mobile device;
in response to blocking access to the feature of the mobile device,
generating, by the processing system, a prompt for an alternate authentication feature associated with the authorized user profile, and
displaying, by the processing system, the prompt via a graphical user interface of the mobile device, wherein the alternate authentication feature comprises a comparison of an input at the mobile device with a defined input known to be associated with the authorized user profile; and
in response to the alternate authentication feature being determined to be completed via the mobile device, unblocking, by the processing system, access to the feature of the mobile device.
19. The method of claim 17 , wherein the reactions correlated to the application over time further comprises a movement of the mobile device associated with the application, a follow-up input on the mobile device associated with the application or a combination thereof.
20. The method of claim 17 , further comprising:
determining, by the processing system, the past motion at the mobile device using a tracking cookie installed on the mobile device.