Utilizing a machine learning model to identify unhealthy online user behavior and to cause healthy physical user behavior

Patent metadata

Field	Value
Publication number	US-12087425-B2
Application number	US-202017247230-A
Country	US
Kind code	B2
Filing date	Dec 4, 2020
Priority date	May 29, 2019
Publication date	Sep 10, 2024
Grant date	Sep 10, 2024

Abstract

A device receives, from a client device, behavior data indicating an action of a user of the client device, and processes the behavior data, with a model, to determine whether the action satisfies a behavior threshold. The device determines preventative actions to perform to prevent the action of the user, when the action is determined to satisfy the behavior threshold, and performs the preventative actions to prevent the action of the user. The device provides, to the client device, a request indicating that the user perform a physical activity before the one or more preventative actions are disabled, and monitors a performance of the physical activity by the user. The device determines whether the user satisfies the performance of the physical activity based on the monitoring, and disables the one or more preventative actions when it is determined that the user satisfies the performance of the physical activity.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a device, historical behavior data indicating actions performed via client devices,
wherein the actions are associated with online activity comprising at least one of:
browsing the Internet,
making online gambling bets,
shopping online, or
making in-game purchases in association with playing a video game;

training, by the device and using the historical behavior data, a machine learning model to determine, based on input behavior data associated with a user:
a type of online activity associated with the input behavior data, and
data indicating whether actions identified by the input behavior data satisfy a behavior threshold associated with the type of online activity;

determining, by the device and using the machine learning model, that the actions identified by the input behavior data satisfy the behavior threshold;
providing, by the device, to a client device associated with the user, and based on the actions identified by the input behavior data satisfying the behavior threshold, an instruction that the user perform one or more exercises;
monitoring, by the device, a performance of the one or more exercises; and
causing, by the device, the client device to prevent the user from performing the actions until the user completes the performance of the one or more exercises.

2. The method of claim 1 , wherein training the machine learning model comprises:
separating the historical behavior data into a training set, a validation set, and a test set;
training the machine learning model using the training set;
validating results of training the machine learning model using the validation set; and
testing the machine learning model using the test set.

3. The method of claim 1 , wherein the behavior threshold is based on the type of online activity.

4. The method of claim 1 , wherein the historical behavior data includes sensor data associated with the client devices, the sensor data including location data indicating a location of at least one of the client devices.

5. The method of claim 1 , wherein the historical behavior data is received from a monitoring application installed on the client devices.

6. The method of claim 1 , wherein the historical behavior data includes sensor data associated with the client devices, the sensor data including a heartrate of the user.

7. The method of claim 1 , wherein a client device of the client devices is a stationary client device, and
wherein a subset of other client devices of the client devices are mobile client devices that are related to the client device via an application running on the client device.

8. A device, comprising:
one or more memories; and
one or more processors, communicatively coupled to the one or more memories, configured to:
receive historical behavior data indicating actions performed via client devices,
wherein the actions are associated with online activity comprising at least one of:
browsing the Internet,
making online gambling bets,
shopping online, or
making in-game purchases in association with playing a video game;

train, using the historical behavior data, a machine learning model to determine, based on input behavior data associated with a user:
a type of online activity associated with the input behavior data, and
data indicating whether actions identified by the input behavior data satisfy a behavior threshold associated with the type of online activity;

determine, using the machine learning model, that the actions identified by the input behavior data satisfy the behavior threshold;
provide, to a client device associated with the user and based on the actions identified by the input behavior data satisfying the behavior threshold, an instruction that the user perform one or more exercises;
monitor a performance of the one or more exercises; and
cause the client device to prevent the user from performing the actions until the user completes the performance of the one or more exercises for a particular time period.

9. The device of claim 8 , wherein the one or more processors, when training the machine learning model, are configured to:
separate the historical behavior data into a training set, a validation set, and a test set;
train the machine learning model using the training set;
validate results of training the machine learning model using the validation set; and
test the machine learning model using the test set.

10. The device of claim 8 , wherein the behavior threshold is based on the type of online spending activity.

11. The device of claim 8 , wherein the historical behavior data includes sensor data associated with the client devices, the sensor data including location data indicating a location of at least one of the client devices.

12. The device of claim 8 , wherein the historical behavior data is received from a monitoring application installed on the client devices.

13. The device of claim 8 , wherein the historical behavior data includes sensor data associated with the client devices, the sensor data including a heartrate of the user.

14. The device of claim 8 , wherein a client device of the client devices is a stationary client device, and
wherein a subset of other client devices of the client devices are mobile client devices that are related to the client device via an application running on the client device.

15. A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:
one or more instructions that, when executed by one or more processors of a device, cause the device to:
receive historical behavior data indicating actions performed via client devices,
wherein the actions are associated with online activity comprising at least one of:
browsing the Internet,
making online gambling bets,
shopping online, or
making in-game purchases in association with playing a video game;

train, using the historical behavior data, a machine learning model to determine, based on input behavior data associated with a user:
a type of online activity associated with the input behavior data, and
data indicating whether actions identified by the input behavior data satisfy a behavior threshold associated with the type of online activity;

determine, using the machine learning model, that the actions identified by the input behavior data satisfy the behavior threshold;
provide, to a client device associated with the user and based on the actions identified by the input behavior data satisfying the behavior threshold, an instruction that the user perform one or more exercises;
monitor a heart rate of the user in connection with a performance of the one or more exercises; and
cause the client device to prevent the user from performing the actions until the user completes the performance of the one or more exercises.

16. The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, that cause the device to train the machine learning model, cause the device to:
separate the historical behavior data into a training set, a validation set, and a test set;
train the machine learning model using the training set;
validate results of training the machine learning model using the validation set; and
test the machine learning model using the test set.

17. The non-transitory computer-readable medium of claim 15 , wherein the behavior threshold is based on the type of online activity.

18. The non-transitory computer-readable medium of claim 15 , wherein the hist

Assignees

• Capital One Services Llc

Inventors

• Edwards Joshua
• Mossoba Michael
• Benkreira Abdelkadar M'Hamed

Classifications

• G06N20/00

  Machine learning · CPC title

• H04L63/101

  Access control lists [ACL] · CPC title

• G16H20/30 Primary

  relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G16H40/67

  for remote operation · CPC title