Determining recommended items for a shopping list

Patent metadata

Field	Value
Publication number	US-11195222-B2
Application number	US-201916725503-A
Country	US
Kind code	B2
Filing date	Dec 23, 2019
Priority date	Dec 23, 2019
Publication date	Dec 7, 2021
Grant date	Dec 7, 2021

Abstract

In an online concierge system, a customer adds items to an online shopping cart. The online concierge system determines key ingredients from the items in the online shopping cart by mapping the items to generic items and removing non-ingredient items and staple items. The online concierge system retrieves recipes including at least one of the key ingredients. The online concierge system determines complementary ingredients based on the other ingredients in the recipes and calculates co-occurrence scores for the complementary ingredients. Using the co-occurrence scores, the online concierge system ranks the complementary ingredients and sends for display a subset of the complementary ingredients as recommended items.

First claim

What is claimed is:

1. A method comprising:
receiving, from a client device, a list of items in an online shopping cart associated with a user, wherein each item in the list is stored in an inventory of items available for the user to purchase, each item in the inventory associated with a user interface widget for adding the item to the online shopping cart;
determining a set of key ingredients from the list of items by:
mapping each item of the list of items to a generic item by applying a machine-learned model to each item, wherein the machine-learned model analyzes a name of the item to determine an associated generic item; and
removing staple items from the mapped generic items, wherein staple items are items commonly used in recipes;

retrieving recipes that contain one or more of the key ingredients;
determining a set of complementary ingredients from the recipes, wherein each complementary ingredient is different from the key ingredients;
scoring each of the set of complementary ingredients based on a frequency of co-occurrence of the complementary ingredient and each of the one or more key ingredients in the recipes;
ranking user interface widgets associated with each of the complementary ingredients based on the scoring, wherein user interface widgets of complementary ingredients that co-occur with one or more of the key ingredients in the recipes are ranked higher than user interface widgets of complementary ingredients that do not co-occur with one or more of the key ingredients in the recipes;
selecting one or more of the highest ranked user interface widgets to recommend to the user;
sending, for display via a user interface presented at the client device, the selected user interface widgets;
receiving, from the client device, a selection of a user interface widget; and
adding an item associated with the user interface widget to the list of items in the online shopping cart.

2. The method of claim 1 , further comprising:
removing non-ingredients from the mapped generic items, wherein non-ingredients include premade foods.

3. The method of claim 1 , wherein the score for each complementary ingredient is inversely related to the number of the recipes including the complementary ingredient.

4. The method of claim 1 , further comprising:
sending, for display via the user interface at the client device, a subset of recipes including the highest ranked complementary ingredient.

5. The method of claim 1 , wherein the complementary ingredients are further scored based on a frequency of occurrence in an online shopping cart associated with the user of the client device.

6. The method of claim 1 , wherein the score for each complementary ingredient is based on a first number of the recipes including the complementary ingredient, a second number of the recipes including a key ingredient of the key ingredients, and a third number of recipes including the complementary ingredient and the key ingredient.

7. The method of claim 1 , further comprising:
receiving a request from the client device to purchase the list of items in the online shopping cart.

8. A non-transitory computer-readable storage medium comprising instructions executable by a processor, the instructions comprising:
instructions for receiving, from a client device, a list of items in an online shopping cart associated with a user, wherein each item in the list is stored in an inventory of items available for the user to purchase, each item in the inventory associated with a user interface widget for adding the item to the online shopping cart;
instructions for determining key ingredients from the list of items, the instructions further comprising:
instructions for mapping each item of the list of items to a generic item by applying a machine-learned model to each item, wherein the machine-learned model analyzes a name of the item to determine an associated generic item; and
instructions for removing staple items from the mapped generic items, wherein staple items are items commonly used in recipes;

instructions for retrieving recipes that contain one or more of the key ingredients;
instructions for determining a set of complementary ingredients from the recipes, wherein each complementary ingredient is different from the key ingredients;
instructions for scoring each of the set of complementary ingredients based on frequency of co-occurrence of the complementary ingredient and each of the one or more key ingredients in the recipes;
instructions for ranking user interface widgets associated with each of the complementary ingredients based on the scoring, wherein user interface widgets of complementary ingredients that co-occur with one or more of the key ingredients in the recipes are ranked higher than user interface widgets of complementary ingredients that do not co-occur with one or more of the key ingredients in the recipes;
instructions for selecting one or more of the highest ranked user interface widgets to recommend to the user;
instructions for sending, for display via a user interface presented at the client device, the selected user interface widgets;
instructions for receiving, from the client device, a selection of a user interface widget; and
instructions for adding an item associated with the user interface widget to the list of items in the online shopping cart.

9. The non-transitory computer-readable storage medium of claim 8 , the instructions further comprising:
instructions for removing non-ingredients from the mapped generic items, wherein non-ingredients include premade foods.

10. The non-transitory computer-readable storage medium of claim 8 , wherein the score for each complementary ingredient is the inverse of the number of the recipes including the complementary ingredient.

11. The non-transitory computer-readable storage medium of claim 8 , the instructions further comprising:
sending, for display via the user interface at the client device, a subset of recipes including the highest ranked complementary ingredient.

12. The non-transitory computer-readable storage medium of claim 8 , wherein the complementary ingredients are further scored based on frequency of occurrence in an online shopping cart associated with the user of the client device.

13. A computer system comprising:
a computer processor; and
a non-transitory computer-readable storage medium storage instructions that when executed by the computer processor perform actions comprising:
receiving, from a client device, a list of items in an online shopping cart associated with a user, wherein each item in the list is stored in an inventory of items available for the user to purchase, each item in the inventory associated with a user interface widget for adding the item to the online shopping cart;

determining key ingredients from the list of items by:
mapping each item of the list of items to a generic item by applying a machine-learned model to each item, wherein the machine-learned model analyzes a name of the item to determine an associated generic item; and
removing staple items from the mapped generic items, wherein staple items are items commonly used in recipes;
retrieving recipes that contain one or more of the key ingredients;
determining a set of complementary ingredients from the recipes, wherein each complementary ingredient is different from the key ingredients;
scoring each of the set of complementary ingredients based on a frequency of co-occurrence of the complementary ingredient and each of the one or more key ingredients in the recipes;
ranking user interface widgets associated with each of the complementary ingredients based on the scoring,

Assignees

• Maplebear Inc

Inventors

• Archak Shrikar
• Rao Sharath

Classifications

• G06Q30/0631 Primary

  Recommending goods or services · CPC title

• G06F16/248

  Presentation of query results · CPC title

• G06Q30/0641 Primary

  utilising user interfaces specially adapted for shopping · CPC title

• G06Q30/0633

  Managing shopping lists, e.g. compiling or processing purchase lists (shipping orders G06Q10/083; order filling G06Q10/087) · CPC title

• G06F16/24578

  using ranking · CPC title