Fashion matching algorithm solution

• US 8,249,941 B2
• Filed: 01/12/2012
• Issued: 08/21/2012
• Est. Priority Date: 07/31/2007
• Status: Active Grant

• Current Assignee: Style D.U. Jour
• Original Assignee: Style D.U. Jour
• Inventors: Saul, Justin, Dantsker, Arkadiy
• Primary Examiner(s): Smith, Jeffrey A
• Assistant Examiner(s): STOPP, COURTNEY L
• Application Number:
  US13/349,464
• Publication Number:
  US 20120116917A1
• Time in Patent Office: 222 Days
• Field of Search: 705/26.1, 705/27.1, 705/26.7, 705/27.2
• US Class Current: 705/26.1
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0603 Catalogue ordering
• CPC Class Codes: G06Q 30/0629 for generating comparisons
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: G06Q 30/0643 Graphical representation of items or shoppers

Abstract

In one example embodiment, a system and method is illustrated that includes receiving an item choice including a fashion item. The system and method also includes matching the fashion item with an additional fashion item selected from a style matrix, the matching based upon an attribute. Further, the system and method includes transmitting the additional fashion item as part of an outfit match set. Moreover, the system and method includes selecting the fashion item from a digital closet. The system and method includes processing the fashion item to build an attribute association matrix. In addition, the system and method includes comparing the attribute association matrix and the style matrix to determine a match of the fashion item and the additional item.

Claims

1. A fashion matching computer system comprising:

a receiver to receive a fashion item choice;

a memory storing a first fashion item, attribute-association matrix and a second fashion item, attribute-association matrix;

a matching engine to match the fashion item with an additional fashion item based on an attribute of the fashion item and a style matrix generated through a threshold function that finds a difference between the first attribute association matrix and the second attribute association matrix, where this difference is less than or equal to a threshold value, wherein the style matrix includes items that are selectable using the matching engine; and

a transmitter to output the additional fashion item as part of an outfit.

2. The computer system of claim 1, wherein the memory stores the fashion item in a digital closet;

wherein a processor is to build the first attribute association matrix using the digital closet; and

wherein the matching engine is to compare the first attribute association matrix and the style matrix to determine a match of the fashion item and the additional fashion item.

3. The computer system of claim 1, wherein the style matrix is generated from the first attribute association matrix that includes the fashion item.

4. The computer system of claim 1, wherein the first attribute association matrix and second attribute association matrix are used in an input layer of a neural network, and the style matrix is used in a computational layer of the neural network.

5. The computer system of claim 1, further comprising a filter to filter the fashion item and the additional fashion item by a color-to-color association.

6. The computer system of claim 1, further comprising a filter to filter the fashion item and the additional fashion item by a price-to-price association.

7. The computer system of claim 6, wherein the filtering includes a weighted, neural network link.

8. The computer system of claim 1, further comprising a filter to filter the fashion item and the additional fashion item by a fabric-to-fabric association.

9. The computer system of claim 1, wherein the matching engine is to compare an image of the fashion item and an additional image of the additional fashion item based upon an image position relative to an area of a projection and to transform the images into a same position.

10. The computer system of claim 1, wherein the transmitter is to generate a prompt to purchase the additional fashion item.

11. The computer system of claim 1, wherein the memory includes an identifier key for data entries in the first attribute association matrix and the second attribute association.

12. The computer system of claim 1, further comprising a filter to filter the fashion item and the additional fashion item by a brand-to-brand association.

13. The computer system of claim 1, wherein the matching engine includes a filter to filter the fashion item and the additional fashion item by a color-to-color association;

a price-to-price association; and

a fabric-to-fabric association.

14. A fashion matching computer system comprising:

a receiver to receive an fashion item choice;

a memory storing a plurality of attribute-association matrices, a style matrix, and a plurality of fashion tables, wherein the plurality of attribute-association matrices are generated using data from the plurality of fashion tables;

a matching engine to match the fashion item choice with an additional fashion item based on an attribute of the fashion item and a style matrix generated through a threshold function that finds a difference between a first attribute association matrix and a second attribute association matrix, where this difference is less than or equal to a threshold value, wherein the matching engine is to select a fashion item from the style matrix; and

a transmitter to output the additional fashion item as part of an outfit.

15. The computer system of claim 14, wherein the receiver receives the fashion item choice from a mobile phone.

16. The computer system of claim 14, wherein the style matrix is generated from the first attribute association matrix that includes the fashion item.

17. The computer system of claim 14, wherein the plurality of fashion tables includes a fashion trends table.

18. The computer system of claim 14, wherein the plurality of fashion tables includes a fashion item type table.

19. The computer system of claim 14, wherein the plurality of fashion tables includes a fashion styles table.

20. The computer system of claim 14, wherein the matching engine includes a filter to filter the fashion item and the additional fashion item based upon a fashion item attribute, the filter is a weighted link in a neural network.

21. The computer system of claim 14, further comprising a neural network that uses plurality of attribute-association matrices to generate the style matrix, wherein the neural network includes a processor to receive multiple inputs and to produce outputs and a first memory coupled to the processor to train.

22. The computer system of claim 14, wherein the receiver receives the fashion item choice from a kiosk.

23. The computer system of claim 14, wherein the receiver receives the fashion item choice from an interactive monitor.

24. The computer system of claim 14, wherein the receiver receives the fashion item choice from an RFID, UPC, a bar code, or combinations thereof.

25. The computer system of claim 14, wherein the plurality of fashion tables includes a fashion trends table;

a fashion item type table; and

a fashion styles table.