Methods and apparatus for automatically providing digital advertisements

Patent metadata

Field	Value
Publication number	US-2020410546-A1
Application number	US-201916455274-A
Country	US
Kind code	A1
Filing date	Jun 27, 2019
Priority date	Jun 27, 2019
Publication date	Dec 31, 2020
Grant date	—

Abstract

This application relates to apparatus and methods for automatically determining and providing digital advertisements to targeted users. In some examples, a computing device receives campaign data identifying items to advertise on a website, and generates campaign user data identifying a user that has engaged all of the items on the website. The computing device may then determine a portion of the users based on a relationship between each user and the campaign user data, and may determine user-item values for each of the items for each user of the portion of users, where each user-item value identifies a relational value between the corresponding user and item. The computing device may then identify one or more of the items to advertise to each user of the portion of users based on the user-item values, and may transmit to a web server an indication of the items to advertise for each user.

First claim

What is claimed is:

1 . A system comprising:
a computing device configured to:
receive campaign data identifying a first plurality of items;
generate first user data based on the first plurality of items, where the first user data identifies a first user that has engaged with the first plurality of items;
obtain attribute data identifying at least one attribute of each of a plurality of users;
determine a portion of the plurality of users based on application of a nearest neighbor algorithm to the first user data and the attribute data;
determine user-item values for each of the first plurality of items for each user of the portion of the plurality of users, wherein each user-item value identifies a relational value between the corresponding user and item;
determine at least one of the first plurality of items for each user of the portion of the plurality of users based on the user-item values; and
transmit an indication of the at least one of the first plurality of items for each user of the portion of the plurality of users.

2 . The system of claim 1 , wherein the computing device is configured to:
receive session data identifying engagement of a second plurality of items for each of the plurality of users; determine at least one attribute value for at least one attribute of the second plurality of items for each of the plurality of the users; generate user representation data for each user of the plurality of users based on application of a neural network to the at least one attribute value and the attribute data corresponding to each user of the plurality of users; and determine the portion of the plurality of users based on the user representation data.

3 . The system of claim 2 , wherein the neural network is a stacked denoising auto-encoder.

4 . The system of claim 2 wherein the computing device is configured to:
determine a common attribute of the second plurality of items for each of the plurality of users; and
determine the at least one attribute value for the at least one attribute of the second plurality of items for each of the plurality of the users based on the common attribute.

5 . The system of claim 4 , wherein attribute values for each of the common attributes of the second plurality of items for each of the plurality of users are averaged.

6 . The system of claim 1 , wherein the nearest neighbor algorithm is an approximate nearest neighbor algorithm.

7 . The system of claim 1 , wherein the computing device is configured to determine the user-item values based on application of a matrix decomposition algorithm.

8 . The system of claim 1 , wherein the computing device is configured to determine the user-item values based on application of a recommender system.

9 . The system of claim 1 , wherein the computing device is configured to:
determine a ranking of the first plurality of items for each user of the portion of the plurality of users based on the user-item values corresponding to each user; and determine a minimum number of items to advertise to each user of the portion of the plurality of users based on the minimum number of the first plurality of items ranked highest.

10 . The system of claim 1 comprising:
a second computing device configured to:
receive the indication of the at least one of the first plurality of items for each user of the portion of the plurality of users;
determine that a first user of the portion of the plurality of users is browsing a website; and
cause the display of an image of the at least one of the first plurality of items for the first user on the website.

11 . A method comprising:
receiving campaign data identifying a first plurality of items; generating first user data based on the first plurality of items, where the first user data identifies a first user that has engaged with the first plurality of items; obtaining attribute data identifying at least one attribute of each of a plurality of users; determining a portion of the plurality of users based on application of a nearest neighbor algorithm to the first user data and the attribute data; determining user-item values for each of the first plurality of items for each user of the portion of the plurality of users, wherein each user-item value identifies a relational value between the corresponding user and item; determining at least one of the first plurality of items for each user of the portion of the plurality of users based on the user-item values; and transmitting an indication of the at least one of the first plurality of items for each user of the portion of the plurality of users.

12 . The method of claim 11 comprising:
receiving session data identifying engagement of a second plurality of items for the plurality of users;
determining at least one attribute value for at least one attribute of the second plurality of items for each of the plurality of the users;
generating user representation data for each user of the plurality of users based on application of a neural network to the at least one attribute value and the attribute data corresponding to each user of the plurality of users; and
determining the portion of the plurality of users based on the user representation data.

13 . The method of claim 12 further comprising:
determining a common attribute of the second plurality of items for each of the plurality of users; and
determining the at least one attribute value for the at least one attribute of the second plurality of items for each of the plurality of the users based on the common attribute.

14 . The method of claim 11 further comprising determining the user-item values based on application of a matrix decomposition algorithm.

15 . The method of claim 11 further comprising:
determining a ranking of the first plurality of items for each user of the portion of the plurality of users based on the user-item values corresponding to each user; and
determining a minimum number of items to advertise to each user of the portion of the plurality of users based on the minimum number of the first plurality of items ranked highest.

16 . A non-transitory computer readable medium having instructions stored thereon, wherein the instructions, when executed by at least one processor, cause a device to perform operations comprising:
receiving campaign data identifying a first plurality of items; determining first user data based on the first plurality of items, where the first user data identifies a first user that has engaged with the first plurality of items; obtaining attribute data identifying at least one attribute of each of a plurality of users; determining a portion of the plurality of users based on application of a nearest neighbor algorithm to the first user data and the attribute data; determining user-item values for each of the first plurality of items for each user of the portion of the plurality of users, wherein each user-item value identifies a relational value between the corresponding user and item; determining at least one of the first plurality of items for each user of the portion of the plurality of users based on the user-item values; and transmitting an indication of the at least one of the first plurality of items for each user of the portion of the plurality of users.

17 . The non-transitory computer readable medium of claim 16 further comprising instructions stored thereon that, when executed by at least one processor, further cause the device to perform operations comprising:
receiving session data identifying engagement of a second pl

Assignees

• Walmart Apollo Llc

Inventors

• Arora Yokila
• Mantha Aditya
• Kanumala Praveenkumar
• Guo Stephen Dean
• Achan Kannan

Classifications

• G06N3/045

  Combinations of networks · CPC title

• G06N3/0455

  Auto-encoder networks; Encoder-decoder networks · CPC title

• G06N3/088

  Non-supervised learning, e.g. competitive learning · CPC title

• G06Q30/0277

  Online advertisement · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title