	US 12,243,008 B2
	Suggesting a recipe to a customer of an online concierge system based on items likely to be available
Karuna Ahuja, San Francisco, CA (US); Girija Narlikar, Palo Alto, CA (US); Sneha Chandrababu, Sammamish, WA (US); Gowri Rajeev, San Francisco, CA (US); Lan Wang, Mountain View, CA (US); Chakshu Ahuja, San Jose, CA (US); and Sonal Jain, Sunnyvale, CA (US)
Assigned to Maplebear Inc., San Francisco, CA (US)
Filed by Maplebear Inc., San Francisco, CA (US)
Filed on Oct. 31, 2022, as Appl. No. 17/977,734.
Prior Publication US 2024/0144173 A1, May 2, 2024
Int. Cl. G06Q 10/00 (2023.01); G06K 7/10 (2006.01); G06K 7/14 (2006.01); G06Q 10/087 (2023.01); G06Q 30/0202 (2023.01); G06Q 30/0601 (2023.01)

CPC G06Q 10/087 (2013.01) [G06K 7/10366 (2013.01); G06K 7/1417 (2013.01); G06Q 30/0202 (2013.01); G06Q 30/0623 (2013.01)]

18 Claims

1. A method, performed at a computer system comprising one or more processors and a computer-readable medium, the method comprising:

detecting a set of acquired items associated with a customer of an online concierge system, wherein the set of acquired items is included among an inventory of the customer;

identifying one or more candidate available items from the set of acquired items based at least in part on one or more of: a predicted perishability of each acquired item and a predicted amount of each acquired item that has been used, wherein identifying one or more candidate available items comprises:

accessing a machine learning model that is trained to predict a likelihood that an item is available, wherein the machine learning model is trained by:

receiving a plurality of attributes associated with a plurality of items included among one or more inventories of one or more retailer locations,

receiving, for each item of the plurality of items, a label indicating an availability of the item,

training the machine learning model based at least in part on the plurality of attributes and the label for each of the plurality of items,

applying the machine learning model to each of the plurality of items to determine a difference between the label and the predicted likelihood of the respective item,

updating the label of the respective item based on the determined difference, and

updating the machine learning model for each of the plurality of items using the updated labels; and

applying the machine learning model to a plurality of attributes of each acquired item of the set of acquired items to predict the likelihood that each acquired item is available;

retrieving a plurality of recipes from a recipe data store, wherein each recipe comprises one or more ingredients;

matching the one or more candidate available items with a set of recipes based at least in part on the one or more ingredients of each recipe and the one or more candidate available items;

identifying a set of remaining items for each recipe of the set of recipes, wherein each remaining item corresponds to an ingredient that is not included among the one or more candidate available items;

retrieving a set of attributes associated with the customer and the set of recipes;

computing a suggestion score for each recipe of the set of recipes based at least in part on the set of attributes associated with the customer and the set of recipes;

ranking the set of recipes based at least in part on the suggestion score for each recipe;

selecting, from the set of recipes, one or more recipes for suggesting to the customer based at least in part on the ranking; and

sending, for display to a customer client device associated with the customer, the one or more recipes and the set of remaining items identified for each of the one or more recipes.