Systems and methods for altering a graphical user interface (GUI) based on affinity and repurchase intent

Patent metadata

Field	Value
Publication number	US-12406282-B2
Application number	US-202318102239-A
Country	US
Kind code	B2
Filing date	Jan 27, 2023
Priority date	Jan 28, 2022
Publication date	Sep 2, 2025
Grant date	Sep 2, 2025

Abstract

Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform: receiving in-session user activity entered into on an initial graphical user interface (GUI) from a user electronic device of a user; pre-processing the in-session user activity to determine one or more intents of the in-session user activity; comparing the one or more intents of the in-session user activity with one or more complementary intents; and coordinating displaying a complimentary GUI on the user device of the user based on the one or more complementary intents. Other embodiments are disclosed herein.

First claim

What is claimed is:

1. A system comprising:
a processor; and
a non-transitory computer-readable medium storing computing instructions that, when executed on the processor, cause the processor to perform:
receiving, from an electronic device of a user, in-session user activity on a graphical user interface (GUI) of a website displayed to the user via the electronic device;
selectively aggregating the in-session user activity by sorting the in-session user activity into one or more groups;
predicting one or more product type intents of the user based on the one or more groups and using one or more predictive algorithms;
post-processing the one or more product type intents to filter out one or more non-relevant product type intents from the one or more product type intents to create filtered product type intents; and
modifying the GUI to display the filtered product type intents to the user, wherein the one or more predictive algorithms are iteratively trained in real time as data is added to a training data set, wherein the data added to the training data set includes dynamic user interactions with the GUI, as modified, during a user session.

2. The system of claim 1 , the computing instructions further cause the processor to perform receiving historical user activity associated with the user, the historical user activity corresponding to historical user activity on the website and historical brick-and-mortar activity at a store having the website.

3. The system of claim 2 , wherein the selectively aggregating the in-session user activity by the sorting the in-session user activity into the one or more groups further comprises weighting the in-session user activity and weighting the historical user activity.

4. The system of claim 3 , wherein the selectively aggregating the in-session user activity by the sorting the in-session user activity into the one or more groups further comprises altering at least a portion of the in-session user activity or the historical user activity based on at least one of: i) features corresponding to purchase patterns of the user, ii) price based features, iii) a number of items in a product type, iv) a number of item groups in a product type, v) a number of unique purchases and repeated purchases in a product type or vi) the in-session user activity, as weighted, and the historical user activity, as weighted.

5. The system of claim 1 , wherein the predicting the one or more product type intents of the user based on the one or more groups using the one or more predictive algorithms further comprises:
converting the in-session user activity into vector format;
incrementing a count in an activity database for a product type intent for the user when no purchase information is present; and
decreasing the count in the activity database when the product type intent for the user has recent purchase information present.

6. The system of claim 1 , wherein the predicting the one or more product type intents of the user based on the one or more groups using the one or more predictive algorithms further comprises applying a multi-class classification algorithm to predict one or more probabilities that the user exhibits the one or more product type intents to perform one or more actions on the GUI.

7. The system of claim 1 , wherein the predicting the one or more product type intents of the user based on the one or more groups using the one or more predictive algorithms further comprises training the one or more predictive algorithms by estimating internal parameters of a model configured to determine the one or more product type intents of the user while interacting with the GUI.

8. The system of claim 7 , wherein the one or more predictive algorithms use an equation comprising:

P

Purch


Pti

=

{

0
,

if
⁢

X
⁢

is
⁢

null
⁢

vector

1

(

1
-

e

-
W

*

X
T

)

,

otherwise

wherein P comprises a probability of the user having a product type intent, X comprises one or more user features, and W comprises one or more learnt weights of a model probability for a complementary action space.

9. The system of claim 1 , wherein the post-processing the one or more product type intents further comprises filtering out the one or more non-relevant product type intents that are below a predetermined likelihood to be performed by the user more than once in a predetermined amount of time.

10. The system of claim 1 , wherein the post-processing the one or more product type intents to filter out the one or more non-relevant product type intents further comprises re-ranking the filtered product type intents using a dot product of the filtered product type intents.

11. A method implemented via execution of computing instructions configured to run at a processor and configured to be stored at a non-transitory computer-readable medium, the method comprising:
receiving, from an electronic device of a user, in-session user activity on a graphical user interface (GUI) of a website displayed to the user via the electronic device;
selectively aggregating the in-session user activity by sorting the in-session user activity into one or more groups;
predicting one or more product type intents of the user based on the one or more groups and using one or more predictive algorithms;
post-processing the one or more product type intents to filter out one or more non-relevant product type intents from the one or more product type intents to create filtered product type intents; and
modifying the GUI to display the filtered product type intents to the user, wherein the one or more predictive algorithms are iteratively trained in real time as data is added to a training data set, wherei

Assignees

• Walmart Apollo Llc

Inventors

• Bajaj Ahsaas
• Cerekovic Aleksandra
• Korpeoglu Evren
• Achan Kannan
• Subramaniam Sinduja

Classifications

• G06F16/9577

  Optimising the visualization of content, e.g. distillation of HTML documents · CPC title

• G06Q30/0201 Primary

  Market modelling; Market analysis; Collecting market data · CPC title

• G06Q30/0202

  Market predictions or forecasting for commercial activities · CPC title

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06Q30/0643

  graphically representing goods, e.g. 3D product representation · CPC title