Granted Patent B2

US 12,321,832 · App. 18/400,156 · Granted Jun 3, 2025

Inventors: Vahid Jalalibarsari (Sunnyvale, CA); Wei Shen (Danville, CA)

Assignee: WALMART APOLLO, LLC

G06N20/00G06F17/18G06N7/00G06Q30/0203

Abstract

Claims (68)

1. A system comprising:

one or more processors; and

one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform functions comprising:

calculating a first user propensity score to take first actions and a second user propensity score to take second actions based on at least one feature vector of historical data of a user;

using the first user propensity score to place the user into a first segment;

using the second user propensity score to place the user into a second segment different than the first segment; and

facilitating a display of one or more selectable elements on a GUI of the user based on the first segment and the second segment.

2. The system of claim 1 , wherein the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform functions comprising:

collecting the historical data of the user;

converting the historical data of the user into the at least one feature vector;

calculating the second user propensity score using the at least one feature vector, wherein the second user propensity score represents a different user propensity than the first user propensity score;

normalizing the first user propensity score; and

normalizing the second user propensity score.

3. The system of claim 2 , wherein normalizing the first user propensity score comprises using a prior correction technique.

4. The system of claim 1 , wherein calculating the first user propensity score comprises:

using a normal model to calculate the first user propensity score for a broader first segment of users than a strict model.

5. The system of claim 4 , wherein the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform functions comprising:

training the normal model using labeled training data from a plurality of metalabels; and

training the strict model using labeled training data from one metalabel.

6. The system of claim 4 , wherein the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform functions comprising:

filtering a set of users into a subset of the users of both the first segment and the second segment; and

delivering, using the strict model, messages to the subset of the users, wherein the messages comprise information about a metalabel.

7. The system of claim 1 , wherein calculating the first user propensity score for the user based on the at least one feature vector comprises:

iteratively determining, using a machine learning algorithm, equations for calculating probabilities of the user.

8. The system of claim 7 , wherein the machine learning algorithm comprises a logistic regression model.

9. The system of claim 1 , wherein the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform functions comprising:

converting the at least one feature vector into a sparse representation of the at least one feature vector; and

storing the sparse representation of the at least one feature vector in a database.

10. The system of claim 1 , wherein the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform functions comprising:

determining when the user interacted with a message;

determining when the user ignored the message; and

adding data to the historical data, wherein the data comprises a date and a time when the user interacted with the message or ignored the message.

11. A method being implemented via execution of computing instructions configured to run on one or more processors and stored at non-transitory computer-readable media, the method comprising:

calculating a first user propensity score to take first actions and a second user propensity score to take second actions based on at least one feature vector of historical data of a user;

using the first user propensity score to place the user into a first segment;

using the second user propensity score to place the user into a second segment different than the first segment; and

facilitating a display of one or more selectable elements on a GUI of the user based on the first segment and the second segment.

12. The method of claim 11 further comprising:

collecting the historical data of the user;

converting the historical data of the user into the at least one feature vector;

calculating the second user propensity score using the at least one feature vector, wherein the second user propensity score represents a different user propensity than the first user propensity score;

normalizing the first user propensity score; and

normalizing the second user propensity score.

13. The method of claim 12 , wherein normalizing the first user propensity score comprises using a prior correction technique.

14. The method of claim 11 , wherein calculating the first user propensity score comprises:

using a normal model to calculate the first user propensity score for a broader first segment of users than a strict model.

15. The method of claim 14 further comprising:

training the normal model using labeled training data from a plurality of metalabels; and

training the strict model using labeled training data from one metalabel.

16. The method of claim 14 further comprising:

(a)

filtering a set of users into a subset of the users of both the first segment and the second segment; and

delivering, using the strict model, messages to the subset of the users, wherein the messages comprise information about a metalabel; or

(b)

determining when the user interacted with a message;

determining when the user ignored the message; and

adding data to the historical data, wherein the data comprises a date and a time when the user interacted with the message or ignored the message.

17. The method of claim 11 , wherein calculating the first user propensity score for the user based on the at least one feature vector comprises:

iteratively determining, using a machine learning algorithm, equations for calculating probabilities of the user.

18. The method of claim 17 , wherein the machine learning algorithm comprises a logistic regression model.

19. The method of claim 11 further comprising:

converting the at least one feature vector into a sparse representation of the at least one feature vector; and

storing the sparse representation of the at least one feature vector in a database.

20. A non-transitory computer-readable medium storing instructions, wherein the instructions, upon execution by a processor, cause the processor to perform operations comprising:

calculating a first user propensity score to take a first action and a second user propensity score to take a second action based on a feature vector of historical data of a user;

using the first user propensity score to place the user into a first segment;

using the second user propensity score to place the user into a second segment different than the first segment; and

facilitating a display of one or more selectable elements on a GUI of the user based on the first segment and the second segment.

Assignments (1)

Continuity (3)

References Cited (38)