User-preferred item attributes

Patent metadata

Field	Value
Publication number	US-11580585-B1
Application number	US-202016915201-A
Country	US
Kind code	B1
Filing date	Jun 29, 2020
Priority date	Jun 29, 2020
Publication date	Feb 14, 2023
Grant date	Feb 14, 2023

Abstract

Disclosed are one or more embodiments for a unique and personalized experience for a user interacting with an electronic commerce site by identifying user-preferred item attributes using supervised machine learning and presenting items to the user in an arrangement that is based on the identified item attributes. A shopping mission is determined according to user interactions with an electronic commerce site. The shopping mission is applied to an attribute prediction model that is trained to detect user-preferred item attributes for items included the item category and estimate a likelihood that an item containing a particular attribute will be purchased or interacted with during the interactions with the electronic commerce site.

First claim

Therefore, the following is claimed:

1. A system, comprising:
at least one computing device; and
at least one application executable by the at least one computing device, wherein, when executed, the at least one application causes the at least one computing device to at least:
identify a current session associated with a user interacting with an electronic commerce site;
determine a current shopping mission based at least in part on one or more interactions by the user with the electronic commerce site during the session in response to detecting a sequence of interaction events in the current session;
identify an item category associated with the current shopping mission, the sequence of interaction events being clustered for the item category in the current shopping mission;
apply one or more event features of the current shopping mission to an attribute preference model trained for the item category;
determine one or more user-preferred item attributes associated with the current session based at least in part on an output of the attribute preference model, the one or more user-preferred item attributes being determined without a user identifying at least one of the one or more user-preferred item attributes;
identify a plurality of items to present during the current session based at least in part on the one or more user-preferred item attributes and the item category;
generate a user interface comprising one or more selectable refinement components corresponding to the one or more user-preferred item attributes and a plurality of item listings corresponding to the plurality of items; and
transmit the user interface to a client device associated with the user.

2. The system of claim 1 , wherein, when executed, the at least one application further causes the at least one computing device to at least determine an order of presentation of the one or more user-preferred attributes based at least in part on a ranking of the one or more user-preferred attributes.

3. The system of claim 2 , wherein, when executed, the at least one application further causes the at least one computing device to at least rank the one or more user-preferred attributes based at least in part on a respective likelihood of selection of an item having a respective user-preferred attribute of the one or more user-preferred attributes during one or more future events of the current shopping mission.

4. The system of claim 1 , wherein, when executed, the at least one application further causes the at least one computing device to at least determine an order of presentation of the plurality of item listings based at least in part on a ranking of the one or more user-preferred attributes.

5. A method, comprising:
determining, by at least one computing device, a current shopping mission of a current user session including one or more interactions with an electronic commerce site in response to detecting a sequence of interaction events in the current session;
identifying, by at least one computing device, an item category associated with the current shopping mission of the current user session, the sequence of interaction events being clustered for the item category in the current shopping mission;
selecting, by the at least one computing device, an attribute prediction model based at least in part on the item category;
determining, by the at least one computing device, a plurality of predicted user-preferred item attributes for the current user session based at least in part on the item category and the selected attribute prediction model, the plurality of predicted user-preferred item attributes being determined without user identification in the current shopping mission; and
generating, by the at least one computing device, a user interface comprising one or more selectable refinement components corresponding to one or more of the plurality of predicted user-preferred item attributes and a listing of items associated with the item category and having one or more of the plurality of predicted user-preferred item attributes.

6. The method of claim 5 , further comprising updating, by the at least one computing device, the user interface based at least in part on a selection of a particular selectable refinement component of the one or more of selectable refinement components, the user interface being updated to refine the listing of items according to a respective attribute associated with the particular selectable refinement component.

7. The method of claim 5 , wherein the plurality of predicted user-preferred item attributes are further determined based at least in part on interaction history data of a user account with respect to the item category, the interaction history data being used as an input to the attribute prediction model.

8. The method of claim 5 , wherein the attribute prediction model estimates a likelihood of an item containing a particular attribute being purchased during the current shopping mission.

9. The method of claim 5 , further comprising:
identifying the listing of items from a pool of recommended items obtained in response to a user request received during the current user session; and
ranking the listing of items based at least in part on the plurality of predicted user-preferred item attributes.

10. The method of claim 9 , wherein the user request corresponds to a search query, an item detail page request, a webpage request for a webpage that includes one or more recommended items, an electronic commerce site home page, or an order summary page.

11. The method of claim 5 , wherein the attribute prediction model is selected from a plurality of attribute prediction models based at least in part on at least one of the current shopping mission or user interaction history.

12. The method of claim 5 , wherein the current shopping mission corresponds to a sequence of events associated with the user account interacting with the electronic commerce site.

13. A system, comprising:
at least one computing device; and
at least one application executable in the at least one computing device, wherein when executed the at least one application causes the at least one computing device to at least:
identify a current session associated with a client device interacting with an electronic commerce site;
determining a current shopping mission in response to detecting a sequence of interaction events in the current session;
determine an item category associated with a current shopping mission of the client device interacting with the electronic commerce site, the sequence of interaction events being clustered for the item category in the current shopping mission;
identify one or more user-preferred attributes based at least in part on the current shopping mission and an attribute prediction model trained for the item category, the one or more user-preferred attributes being identified without a prior identification of at least one of the one or more user-preferred attributes; and
generate a user interface identifying the one or more user-preferred attributes and a listing of items presented in an order based at least in part on a ranking of the one or more user-preferred attributes.

14. The system of claim 13 , wherein, when executed, the at least one application causes the at least one computing device to at least train the attribute prediction model based at least in part on a plurality of prior shopping missions.

15. The system of claim 14 , wherein the plurality of prior shopping missions are associated with interaction history data obtained from a plurality of client devices interacting with the electronic commerc

Assignees

• Amazon Tech Inc

Inventors

• Rajana Sneha
• Allison Ben
• Zhao Xiaonan
• Varia Pranav
• Yoon Sungro
• Scott Charles Lawrence
• Lloyd Daniel

Classifications

• G06Q30/0641 Primary

  utilising user interfaces specially adapted for shopping · CPC title

• G06Q30/0627

  by specifying product or service characteristics, e.g. product dimensions · CPC title

• G06Q30/0202

  Market predictions or forecasting for commercial activities · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title