Systems, methods, and devices for measuring similarity of and generating recommendations for unique items

• US 10,643,265 B2
• Filed: 06/08/2017
• Issued: 05/05/2020
• Est. Priority Date: 03/07/2013
• Status: Active Grant

• Current Assignee: Vast.com, Inc. (Vroom, Inc.)
• Original Assignee: Vast.com, Inc. (Vroom, Inc.)
• Inventors: Levy, Joshua Howard, Franke, David Wayne
• Primary Examiner(s): Chen, Alan
• Application Number:
  US15/617,927
• Publication Number:
  US 20170372402A1
• Time in Patent Office: 1,062 Days
• Field of Search: None
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

The disclosure herein provides methods, systems, and devices for measuring similarity of and generating recommendations for unique items. A recommendation system for generating recommendations of alternative unique items comprises an items information database, a penalty computation engine, a recommendation compilation engine, and one or more computers, wherein the penalty computation engine comprises a customizations filter, a condition filter, and a dissimilarity penalty calculator.

Claims

1. A computer-based recommendation system for generating recommendations of unique items, the recommendation system comprising:

one or more computer readable storage devices configured to store;

a plurality of computer executable instructions; and

an items information database comprising data relating to a plurality of unique items; and

one or more hardware computer processors in communication with the one or more computer readable storage devices and configured to execute the plurality of computer executable instructions in order to cause the computer system to;

render an electronic user interface for users to interact with listings of unique items;

receive, via the electronic user interface, user data, the user data comprising at least user inputs comprising user-expressed preferences associated with the plurality of unique items;

calculate a status score for each unique item in the plurality of unique items, the status score at least partially based on an estimated preference impact of at least one status attribute associated with that unique item;

calculate a customization score for each unique item in the plurality of unique items, the customization score at least partially based on an estimated preference impact of at least one customization attribute associated with that unique item;

generate a dissimilarity penalty for each unique item in the plurality of unique items by combining at least the status score and the customization score for that unique item, the dissimilarity penalty at least partially based on a magnitude of dissimilarity between the user-expressed preferences and the unique item;

generate a recommendation of unique items by ranking at least a portion of the plurality of the unique items based at least partially on the calculated dissimilarity penalties; and

render a visual representation of the recommendation of unique items via the electronic user interface.

2. The computer-based recommendation system of claim 1, wherein the unique items comprise one of the following types of items:

automobiles, homes, real estate, household goods, customized electronics, customized goods.

3. The computer-based recommendation system of claim 1, wherein the at least one customization attribute describes at least one of the following:

an engine size, a type of material used for an interior of an automobile, a color of an automobile.

4. The computer-based recommendation system of claim 1, wherein the at least one status attribute describes at least one of the following:

a listing price, a geographic location, a type of seller.

5. The computer-based recommendation system of claim 1, wherein the calculation of the customization score comprises a Mahalanobis distance calculation.

6. The computer-based recommendation system of claim 1, wherein the status score is further at least partially based on an estimated preference impact of at least one condition attribute.

7. The computer-based recommendation system of claim 6, wherein the at least one condition attribute describes at least one of the following:

a number of miles an automobile has been driven, whether an automobile'"'"'s title is clean, whether an automobile is new or used, whether an automobile has been in an accident.

8. A computer-implemented method for generating recommendations of unique items, the computer-implemented method comprising:

rendering an electronic user interface for users to interact with listings of unique items;

receiving, via the electronic user interface, user data, the user data comprising at least user inputs comprising user-expressed preferences associated with a plurality of unique items;

calculating, by a computer system, a status score for each unique item in the plurality of unique items, the status score at least partially based on an estimated preference impact of at least one status attribute associated with that unique item;

calculating, by the computer system, a customization score for each unique item in the plurality of unique items, the customization score at least partially based on an estimated preference impact of at least one customization attribute associated with that unique item;

generating, by the computer system, a dissimilarity penalty for each unique item in the plurality of unique items by combining at least the status score and the customization score for that unique item, the dissimilarity penalty at least partially based on a magnitude of dissimilarity between the user-expressed preferences and the unique item;

generating, by the computer system, a recommendation of unique items by ranking at least a portion of the plurality of the unique items based at least partially on the calculated dissimilarity penalties; and

rendering a visual representation of the recommendation of unique items via the electronic user interface,

wherein the computer system comprises a computer processor and electronic memory.

9. The computer-implemented method of claim 8, wherein the unique items comprise one of the following types of items:

automobiles, homes, real estate, household goods, customized electronics, customized goods.

10. The computer-implemented method of claim 8, wherein the at least one customization attribute describes at least one of the following:

an engine size, a type of material used for an interior of an automobile, a color of an automobile.

11. The computer-implemented method of claim 8, wherein the at least one status attribute describes at least one of the following:

a listing price, a geographic location, a type of seller.

12. The computer-implemented method of claim 8, wherein the calculation of the customization score comprises a Mahalanobis distance calculation.

13. The computer-implemented method of claim 8, wherein the status score is further at least partially based on an estimated preference impact of at least one condition attribute.

14. The computer-implemented method of claim 13, wherein the at least one condition attribute describes at least one of the following:

a number of miles an automobile has been driven, whether an automobile'"'"'s title is clean, whether an automobile is new or used, whether an automobile has been in an accident.

15. A computer-readable, non-transitory storage medium having a computer program stored thereon for causing a suitably programmed computer system to process by one or more processors computer-program code by performing a method for generating recommendations of unique items when the computer program is executed on the suitably programmed computer system, the method comprising:

rendering an electronic user interface for users to interact with listings of unique items;

receiving, via the electronic user interface, user data, the user data comprising at least user inputs comprising user-expressed preferences associated with a plurality of unique items;

calculating, by a computer system, a status score for each unique item in the plurality of unique items, the status score at least partially based on an estimated preference impact of at least one status attribute associated with that unique item;

calculating, by the computer system, a customization score for each unique item in the plurality of unique items, the customization score at least partially based on an estimated preference impact of at least one customization attribute associated with that unique item;

generating, by the computer system, a dissimilarity penalty for each unique item in the plurality of unique items by combining at least the status score and the customization score for that unique item, the dissimilarity penalty at least partially based on a magnitude of dissimilarity between the user-expressed preferences and the unique item;

generating, by the computer system, a recommendation of unique items by ranking at least a portion of the plurality of the unique items based at least partially on the calculated dissimilarity penalties; and

rendering a visual representation of the recommendation of unique items via the electronic user interface,

wherein the computer system comprises a computer processor and electronic memory.

16. The computer-readable, non-transitory storage medium of claim 15, wherein the unique items comprise one of the following types of items:

automobiles, homes, real estate, household goods, customized electronics, customized goods.

17. The computer-readable, non-transitory storage medium of claim 15, wherein the at least one customization attribute describes at least one of the following:

an engine size, a type of material used for an interior of an automobile, a color of an automobile.

18. The computer-readable, non-transitory storage medium of claim 15, wherein the at least one status attribute describes at least one of the following:

a listing price, a geographic location, a type of seller.

19. The computer-readable, non-transitory storage medium of claim 15, wherein the status score is further at least partially based on an estimated preference impact of at least one condition attribute.

20. The computer-readable, non-transitory storage medium of claim 19, wherein the at least one condition attribute describes at least one of the following:

a number of miles an automobile has been driven, whether an automobile'"'"'s title is clean, whether an automobile is new or used, whether an automobile has been in an accident.