US 12,354,135 B2
Personalized ranking of promotional items
Haema Nilakanta, Minneapolis, MN (US); Chad Morgan, Sunnyvale, CA (US); Luyen Le, Minneapolis, MN (US); Anick Saha, Bellevue, WA (US); and Kusumakumari Vanteru, Sunnyvale, CA (US)
Assigned to Target Brands, Inc., Minneapolis, MN (US)
Filed by Target Brands, Inc., Minneapolis, MN (US)
Filed on Jun. 6, 2023, as Appl. No. 18/330,234.
Prior Publication US 2024/0412252 A1, Dec. 12, 2024
Int. Cl. G06Q 30/0251 (2023.01)
CPC G06Q 30/0255 (2013.01) [G06Q 30/0269 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A method for dynamically presenting items eligible for a promotion on an e-commerce application of a retailer, the method comprising:
receiving historical transaction data from a transaction dataset associated with a user;
extracting item features from the transaction data;
based on at least the extracted item features, training a machine learning model for generation of pre-computed components associated with the user;
generating, at the trained machine learning model, pre-computed components associated with the user;
receiving interaction data from an interaction data store and item data from an item data store;
based on at least the interaction data and the item data, training the machine learning model for generation of real-time components;
receiving, from a computing device associated with a customer user in a current online shopping session, a request for an offer description page of the promotion;
retrieving items eligible for the promotion and interaction data associated with the current online shopping session;
generating, at the trained machine learning model, real-time components associated with the user;
based on both the generated pre-computed components and the generated real-time components, assigning a relevancy score for each of the items eligible for the promotion in real-time,
wherein the pre-computed components predict a likelihood the customer user will purchase an item based at least in part on profile data of the customer user and on item features,
wherein the real-time components predict a likelihood the customer user will purchase an item based at least in part on the interaction data collected in real-time;
ranking the items eligible for the promotion based at least in part on the assigned relevancy scores; and
generating a user interface for the offer description page presenting the items eligible for the promotion according to the ranking, wherein the offer description page initially displays a subset of the highest ranked items eligible for the promotion.