Recommendation system using a recipe database and co-occurrences of historical item selections

Patent metadata

Field	Value
Publication number	US-2025069126-A1
Application number	US-202318236342-A
Country	US
Kind code	A1
Filing date	Aug 21, 2023
Priority date	Aug 21, 2023
Publication date	Feb 27, 2025
Grant date	—

Abstract

An online system identifies recipes that are most likely to be pertinent to particular users of the system. To do so, the online system uses an association table containing degrees of association between pairs of possible ingredients, identifying degrees of association between the constituent ingredients of various possible recipes and between ingredients from known user personalization data about the user to whom recipes are being recommended. These degrees of association are used to compute a score for each recipe as a whole, with the highest scores indicating the most pertinent recipes for the user in question. The most pertinent recipes, and/or the constituent ingredients of those recipes, are recommended to the user, and the system may additionally aid the user in obtaining the full complement of ingredients for a recommended recipe. The system may also build the association table as a co-occurrence graph of pairs of items that were previously purchased together by users of the system.

First claim

What is claimed is:

1 . A method performed at a computer system comprising a processor and a computer-readable medium, the method comprising:
accessing user personalization data associated with a user, the user personalization data comprising identifiers of a set of items with which the user interacted; for each recipe of a set of recipes:
accessing an association table that stores degrees of association between pairs of items;
for each of a set of items of the recipe, obtaining a degree of association between the item and an item of the user personalization data from the association table;
computing a score for the recipe using the degrees of association;
selecting a recipe from the set of recipes based on the scores computed for the recipes; and causing presentation of the selected recipe by a device associated with the user in a graphical user interface of the device.

2 . The method of claim 1 , wherein the user personalization data comprises identifiers of items currently within a shopping list of the user.

3 . The method of claim 1 , wherein the user personalization data comprises identifiers of items previously purchased by the user.

4 . The method of claim 1 , wherein the user personalization data comprises identifiers of items within a food storage unit of the user.

5 . The method of claim 4 , further comprising:
receiving an image of contents of the food storage unit; and performing image analysis of the image using a machine-learned model to identify the items within the food storage unit.

6 . The method of claim 1 , further comprising:
identifying a set of items currently in a shopping list of the user; identifying items used by the recipe that are not already present in the set of items currently in the shopping list; and causing display to the user of the identified items used by the recipe that are not already present in the set of items currently in the shopping list.

7 . The method of claim 6 , further comprising placing, within the shopping list of the user, at least some of the identified items used by the recipe that are not already present in the set of items currently in the shopping list.

8 . The method of claim 1 , further comprising generating the association table, the generating comprising:
identifying orders of a plurality of users over a given period of time; and computing, as an association degree in the association table for a first item and a second item, a co-occurrence measure for the first item and the second item evaluated over the identified orders of the plurality of users.

9 . The method of claim 8 , wherein the user personalization data comprise identifiers of a set of items that the user has purchased during the given period of time.

10 . The method of claim 8 , wherein the co-occurrence measure for the first item and the second item is defined as (AB*¬A¬B-A¬B*¬AB)/(AB*¬A¬B+A¬B*¬AB), where A represents the first item and B represents the second item.

11 . A non-transitory computer-readable storage medium containing instructions that when executed by one or more processors perform actions comprising:
accessing user personalization data associated with a user, the user personalization data comprising identifiers of a set of items with which the user interacted; for each recipe of a set of recipes:
accessing an association table that stores degrees of association between pairs of items;
for each of a set of items of the recipe, obtaining a degree of association between the item and an item of the user personalization data from the association table;
computing a score for the recipe using the degrees of association;
selecting a recipe from the set of recipes based on the scores computed for the recipes; and causing presentation of the selected recipe by a device associated with the user in a graphical user interface of the device.

12 . The non-transitory computer-readable storage medium of claim 11 , wherein the user personalization data comprise identifiers of items currently within a shopping list of the user.

13 . The non-transitory computer-readable storage medium of claim 11 , wherein the user personalization data comprise identifiers of a set of items that the user has purchased during the given period of time.

14 . The non-transitory computer-readable storage medium of claim 11 , wherein the user personalization data comprise identifiers of items within a food storage unit of the user.

15 . The non-transitory computer-readable storage medium of claim 14 , the actions further comprising:
receiving an image of contents of the food storage unit; and performing image analysis of the image using a machine-learned model to identify the items within the food storage unit.

16 . The non-transitory computer-readable storage medium of claim 11 , the actions further comprising:
identifying a set of items currently in a shopping list of the user; identifying items used by the recipe that are not already present in the set of items currently in the shopping list; and causing display to the user of the identified items used by the recipe that are not already present in the set of items currently in the shopping list.

17 . The non-transitory computer-readable storage medium of claim 16 , the actions further comprising placing, within the shopping list of the user, at least some of the identified items used by the recipe that are not already present in the set of items currently in the shopping list.

18 . The non-transitory computer-readable storage medium of claim 11 , the actions further comprising generating the association table, the generating comprising:
identifying orders of a plurality of users over a given period of time; and computing, as an association degree in the association table for a first item and a second item, a co-occurrence measure for the first item and the second item evaluated over the identified orders of the plurality of users.

19 . The non-transitory computer-readable storage medium of claim 18 , wherein the co-occurrence measure for the first item and the second item is defined as (AB*¬A¬B-A¬B*¬AB)/(AB*¬A¬B+A¬B*¬AB), where A represents the first item and B represents the second item.

20 . A computer system comprising:
one or more computer processors; and a computer-readable storage medium storing instructions that when executed by the one or more computer processors perform actions comprising:
accessing user personalization data associated with a user, the user personalization data comprising identifiers of a set of items with which the user interacted;
for each recipe of a set of recipes:
accessing an association table that stores degrees of association between pairs of items;
for each of a set of items of the recipe, obtaining a degree of association between the item and an item of the user personalization data from the association table;
computing a score for the recipe using the degrees of association;

selecting a recipe from the set of recipes based on the scores computed for the recipes; and
causing presentation of the selected recipe by a device associated with the user in a graphical user interface of the device.

Assignees

• Maplebear Inc

Inventors

• Mesard Madeline
• Mccullough Kevin
• Martin Ryan
• Park Justin

Classifications

• G06F16/2282

  Tablespace storage structures; Management thereof · CPC title

• G06Q10/087

  Inventory or stock management, e.g. order filling, procurement or balancing against orders · CPC title

• G06Q30/0623 Primary

  by investigating goods or services · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title