Systems and methods for altering user interfaces using predicted user activity

Patent metadata

Field	Value
Publication number	US-11710037-B2
Application number	US-202016749116-A
Country	US
Kind code	B2
Filing date	Jan 22, 2020
Priority date	Jan 28, 2019
Publication date	Jul 25, 2023
Grant date	Jul 25, 2023

Abstract

Systems and methods including one or more processing modules and one or more non-transitory storage modules storing computing instructions configured to run on the one or more processing modules and perform acts of: monitoring first activities of a user over a first time period; based on the first activities of the user over the first time period, identifying, using a Markov model, a first probability of the user being in a first state; determining when the first probability is above a first probability predefined threshold; in response to determining when the first probability is above the first probability predefined threshold, automatically customizing first content on a graphical user interface for the first state to create a first graphical user interface on an electronic device of the user while the user is determined to be in the first state; monitoring second activities of the user over a second time period occurring after the first time period and after the user has been determined to be in the first state and before the user has been determined to be in a second state; based on the second activities of the user over the second time period, identifying, using a mixed model different from the Markov model, a second probability that the user has transitioned from the first state into the second state, wherein the second state is related to the first state; determining when the second probability is above a second probability predefined threshold; and in response to determining when the second probability is above the second probability predefined threshold, automatically customizing a second content on the graphical user interface for the second state to create a second graphical user interface on the electronic device user while the user is determined to be in the second state. Other embodiments are disclosed herein.

First claim

What is claimed is:

1. A system comprising:
one or more processors; and
one or more non-transitory computer-readable storage devices storing computing instructions configured to run on the one or more processors and perform acts of:
monitoring first activities of a user over a first time period;
based on the first activities of the user over the first time period, identifying, using a Markov model, a first probability of the user being in a first state;
determining when the first probability is above a first probability predefined threshold;
in response to determining when the first probability is above the first probability predefined threshold, automatically customizing first content on a graphical user interface for the first state to create a first graphical user interface on an electronic device of the user while the user is determined to be in the first state;
monitoring second activities of the user over a second time period occurring after the first time period and after the user has been determined to be in the first state and before the user has been determined to be in a second state;
based on the second activities of the user over the second time period, identifying, using a mixed model different from the Markov model, a second probability that the user has transitioned from the first state into the second state, wherein the second state is related to the first state;
determining when the second probability is above a second probability predefined threshold; and
in response to determining when the second probability is above the second probability predefined threshold, automatically customizing a second content on the graphical user interface for the second state to create a second graphical user interface on the electronic device of the user while the user is determined to be in the second state.

2. The system of claim 1 , wherein the computing instructions are further configured to perform acts of:
training the Markov model to identify the first probability of the user being in the first state, wherein training the Markov model comprises:
identifying one or more binary classifiers, each binary classifier of the one or more binary classifiers independently capable of identifying the first probability of the user being in the first state; or
identifying a multi-class classifier capable of assigning a distribution over the first probability of the user being in the first state.

3. The system of claim 2 , wherein the Markov model is trained on a deep neural network.

4. The system of claim 1 , wherein the mixed model comprises a mix of a Gaussian model and a second Markov model.

5. The system of claim 1 , wherein monitoring the first activities of the user comprises:
gathering information comprising at least one of:
views of an item of a category of items;
cart adds of the item of the category of items;
registry adds of the item of the category of items;
transactions involving the item of the category of items; or
searches for the item of the category of items.

6. The system of claim 1 , wherein automatically customizing the first content on the graphical user interface comprises at least one of:
automatically changing one or more images on the graphical user interface to first images related to the first state;
automatically changing text displayed on the graphical user interface to first text related to the first state; or
automatically altering a layout of the graphical user interface for the first state.

7. The system of claim 1 , wherein the first and second states comprise life events in a sequence of life events of the user.

8. The system of claim 1 , wherein the Markov model comprises:
P(D)=Π t=1 T P(s t , f t (u)), wherein:
P(D) comprises a probability of a user being in the first state;
s t comprises a state at time t; and
f t (u) comprises a set of observed feature of a user u.

9. The system of claim 8 , wherein the Markov model further comprises:
P(D)=Π t=1 T π(s t ; f t (u))ψ(s t , s t+1 ; Δt), wherein ψ(s t , s t+1 ; Δt) comprises a probability of a second state, s t+1 , following a first state, s t , in Δt units of time.

10. The system of claim 1 , wherein:
the mixed model comprises a mix of a Gaussian model and a second Markov model; and
the first and second states comprise life events in a sequence of life events of the user.

11. A method being implemented via execution of computing instructions configured to run at one or more processors and configured to be stored at non-transitory computer-readable media, the method comprising:
monitoring first activities of a user over a first time period;
based on the first activities of the user over the first time period, identifying, using a Markov model, a first probability of the user being in a first state;
determining when the first probability is above a first probability predefined threshold;
in response to determining when the first probability is above the first probability predefined threshold, automatically customizing first content on a graphical user interface for the first state to create a first graphical user interface on an electronic device of the user while the user is determined to be in the first state;
monitoring second activities of the user over a second time period occurring after the first time period and after the user has been determined to be in the first state and before the user has been determined to be in a second state;
based on the second activities of the user over the second time period, identifying, using a mixed model different from the Markov model, a second probability that the user has transitioned from the first state into the second state, wherein the second state is related to the first state;
determining when the second probability is above a second probability predefined threshold; and
in response to determining when the second probability is above the second probability predefined threshold, automatically customizing a second content on the graphical user interface for the second state to create a second graphical user interface on the electronic device of the user while the user is determined to be in the second state.

12. The method of claim 11 further comprising:
training the Markov model to identify the first probability of the user being in the first state, wherein training the Markov model comprises:
identifying one or more binary classifiers, each binary classifier of the one or more binary classifiers independently capable of identifying the first probability of the user being in the first state; or
identifying a multi-class classifier capable of assigning a distribution over the first probability of the user being in the first state.

13. The method of claim 12 , wherein the Markov model is trained on a deep neural network.

14. The method of claim 11 , wherein the mixed model comprises a mix of a Gaussian model and a second Markov model.

15. The method of claim 11 , wherein monitoring the first activities of the user comprises:
gathering information comprising at least one of:
views of an item of a category of items;
cart adds of the item of the category of items;
registry adds of the item of the category of items;
transactions involving the item of the category of items; or
searches for the item of the category of items.

16. The method of claim 11 , wherein automatically customizing the first content on the graphical user interface comprises at least one of:
automatically changing one or more images on the graphical user interface to first images related to the first state;
automatically

Assignees

• Walmart Apollo Llc

Inventors

• Achan Kannan
• Ali Afroza

Classifications

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/08 Primary

  Learning methods · CPC title

• G06F9/451

  Execution arrangements for user interfaces · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title