US 12,548,062 B2Granted

Trained machine learning models for predicting replacement items using expiration dates

AssigneeMaplebear Inc.

InventorsRamasubramanian Balasubramanian, Sharath Rao Karikurve

Life of this patent

Estimating the term…

Loading patent…

Abstract

A specific item is identified to suggest a replacement therefor to a user. A set of candidate replacement items for the specific item is determined. For at least one of the candidate replacement items, an expiration score is determined based on expiration information associated with the item. A replacement score for the candidate replacement item is determined by inputting the determined expiration score as a feature into a machine learning model that is trained using features of historical samples of candidate replacement items suggested as a replacement to users and the replacement suggestion being accepted by the users. One or more of the candidate replacement items is selected based on respective replacement scores as one or more suggested replacement items. A graphical user interface of a client device of the user is caused to display the one or more suggested replacement items as the replacement for the specific item.

Claims20

Dependent claims

A method for suggesting personalized expiration date aware replacements, the method comprising, at a computer system comprising a processor and a computer-readable medium:

receiving, via a graphical user interface of an online concierge system displayed on a client device of a user, an order comprising a plurality of items;
identifying a specific one of the plurality of items to suggest a replacement for the specific item;
determining a set of candidate replacement items for the specific item;
for at least one of the set of candidate replacement items:
determining an expiration score based on expiration information associated with the candidate replacement item;
normalizing the expiration score to compute a normalized expiration factor for the candidate replacement item based on order frequency information associated with the user;
and
determining a replacement score for the candidate replacement item by inputting the normalized expiration factor as a feature into a machine learning model that is trained using features of historical samples of candidate replacement items suggested as a replacement to users and the replacement suggestion being accepted by the users, wherein the replacement score for the candidate replacement item is decreased as a function of the normalized expiration factor, such that a smaller normalized expiration factor produces a smaller replacement score;
selecting one or more of the candidate replacement items based on respective highest replacement scores as one or more suggested replacement items;
receiving, from the user, a confirmation identifying a particular one of the suggested replacement items as a replacement for the specific item;
in response to the confirmation, transmitting to a picker client device:
(i) store-navigation information and (ii) augmented-reality (AR) data configured to overlay guidance within a live camera view of the picker client device to:
(a) navigate a picker to a physical location of the particular one of the suggested replacement items in a retailer location;
and (b) locate and retrieve a corresponding instance of the particular one of the suggested replacement items that has an expiration date matching the expiration information associated with the user-selected replacement item;
receiving, from the picker client device, at least one of:
(i) a scanned item identifier obtained by a barcode scanner, or (ii) image data of an item imaged by a camera, and, based on the scanned item identifier or the image data, verifying that the retrieved item corresponds to the user-selected replacement item;
and
updating an order-fulfillment status to indicate successful retrieval of the user-selected replacement item.

15 more elements

A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to:

receive, via a graphical user interface of an online concierge system displayed on a client device of a user, an order comprising a plurality of items;
identify a specific one of the plurality of items to suggest a replacement for the specific item;
determine a set of candidate replacement items for the specific item;
for at least one of the set of candidate replacement items:
determine an expiration score based on expiration information associated with the candidate replacement item;
normalize the expiration score to compute a normalized expiration factor for the candidate replacement item based on order frequency information associated with the user;
and
determine a replacement score for the candidate replacement item by inputting the normalized expiration factor as a feature into a machine learning model that is trained using features of historical samples of candidate replacement items suggested as a replacement to users and the replacement suggestion being accepted by the users, wherein the replacement score for the candidate replacement item is decreased as a function of the normalized expiration factor, such that a smaller normalized expiration factor produces a smaller replacement score;
select one or more of the candidate replacement items based on respective highest replacement scores as one or more suggested replacement items;
receive, from the user, a confirmation identifying a particular one of the suggested replacement items as a replacement for the specific item;
in response to the confirmation, transmit to a picker client device:
(i) store-navigation information and (ii) augmented-reality (AR) data configured to overlay guidance within a live camera view of the picker client device to:
(a) navigate a picker to a physical location of the particular one of the suggested replacement items in a retailer location;
and (b) locate and retrieve a corresponding instance of the particular one of the suggested replacement items that has an expiration date matching the expiration information associated with the user-selected replacement item;
receive, from the picker client device, at least one of:
(i) a scanned item identifier obtained by a barcode scanner, or (ii) image data of an item imaged by a camera, and, based on the scanned item identifier or the image data, verifying that the retrieved item corresponds to the user-selected replacement item;
and
update an order-fulfillment status to indicate successful retrieval of the user-selected replacement item.

15 more elements

An online concierge system comprising:

one or more processors;
and
memory operatively coupled to the one or more processors and comprising instructions that, when executed by the one or more processors, cause the one or more processors to:
receive, via a graphical user interface of an online concierge system displayed on a client device of a user, an order comprising a plurality of items;
identify a specific one of the plurality of items to suggest a replacement for the specific item;
determine a set of candidate replacement items for the specific item;
for at least one of the set of candidate replacement items:
determine an expiration score based on expiration information associated with the candidate replacement item;
normalize the expiration score to compute a normalized expiration factor for the candidate replacement item based on order frequency information associated with the user;
and
determine a replacement score for the candidate replacement item by inputting the normalized expiration factor as a feature into a machine learning model that is trained using features of historical samples of candidate replacement items suggested as a replacement to users and the replacement suggestion being accepted by the users, wherein the replacement score for the candidate replacement item is decreased as a function of the normalized expiration factor, such that a smaller normalized expiration factor produces a smaller replacement score;
select one or more of the candidate replacement items based on respective highest replacement scores as one or more suggested replacement items;
receive, from the user, a confirmation identifying a particular one of the suggested replacement items as a replacement for the specific item;
in response to the confirmation, transmit to a picker client device:
(i) store-navigation information and (ii) augmented-reality (AR) data configured to overlay guidance within a live camera view of the picker client device to:
(a) navigate a picker to a physical location of the particular one of the suggested replacement items in a retailer location;
and (b) locate and retrieve a corresponding instance of the particular one of the suggested replacement items that has an expiration date matching the expiration information associated with the user-selected replacement item;
receive, from the picker client device, at least one of:
(i) a scanned item identifier obtained by a barcode scanner, or (ii) image data of an item imaged by a camera, and, based on the scanned item identifier or the image data, verifying that the retrieved item corresponds to the user-selected replacement item;
and
update an order-fulfillment status to indicate successful retrieval of the user-selected replacement item.

18 more elements

Dependencies are inferred from the claim text.