Ranking search results based on color

• US 9,460,157 B2
• Filed: 12/28/2012
• Issued: 10/04/2016
• Est. Priority Date: 12/28/2012
• Status: Active Grant

• Current Assignee: Walmart Apollo, LLC (WalMart Inc.)
• Original Assignee: Wal-Mart Stores Texas LLC (WalMart Inc.)
• Inventors: Vuong, Ba-Quy, Gattani, Abhishek, Tholiya, Namrata PramodKumar
• Primary Examiner(s): Morrison, Jay
• Assistant Examiner(s): Hoang, Ken
• Application Number:
  US13/730,765
• Publication Number:
  US 20140188853A1
• Time in Patent Office: 1,376 Days
• Field of Search: 715/738, 715/753, 715/733, 715/226, 715/234, 715/253, 715/719, 715/744, 715/746, 715/747, 707/769, 707/E17.014, 707/722, 707/E17.03, 707/E17.044, 707/706, 707/708, 707/710, 707/731, 707/732, 707/737, 707/753, 707/812, 709/203, 709/206, 709/213, 705/14.43, 705/12, 705/14.23, 705/14.49, 705/14.53, 705/2, 705/26.1, 705/26.5, 705/26.63, 705/30, 705/7.11, 705/7.36
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/9038 Presentation of query results
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]

Abstract

Example systems and methods that rank search results based on color are described. In one implementation, a method receives a product search query from a user and identifies a reference to a color in the product search query. The product search query is executed to identify multiple products. The method further identifies references to the color in a title or a description of each of the multiple products. A ranking of the multiple products is created by increasing the ranking of products that include a reference to the color in the title or the description.

Claims

1. A method comprising:

receiving a product search query from a user;

identifying a reference to a stated color in the product search query by comparing each term in the product search query with a listing of known color terms, wherein the reference to the stated color in the product search query is a keyword in the product search query;

determining one or more additional references of one or more additional colors associated with the stated color in the product search query based on one or more similarity scores each indicating a color distance of a color pair between an individual color of the one or more additional colors and the stated color in the product search query, the one or more similarity scores calculated using a color difference formula;

executing the product search query to identify a plurality of products;

identifying a plurality of references to the stated color contained in a title or a description of multiple of the plurality of products;

calculating a cosine similarity between a search query vector and vectors of identified titles to determine whether the user intended an actual color in the product search query;

creating, using one or more processors, a ranking of the plurality of products by increasing ranking of one or more particular products of the plurality of products that include the reference to the stated color in the title or the description of the one or more particular products and including in the ranking below the one or more particular products multiple similarly colored products of the plurality of products that include the one or more additional references of the one or more additional colors associated with the stated color in the product search query; and

communicating the ranking of the plurality of products to a device operated by the user.

2. The method of claim 1, further comprising causing to display the ranking of the plurality of products to the user.

3. The method of claim 1, wherein the creating the ranking of the plurality of products comprises creating the ranking of the plurality of products without using the reference to the stated color in the product search query in response to a determination that the user did not intend the actual color.

4. The method of claim 1, wherein analyzing of the product search query to determine whether the user intended the actual color includes calculating the cosine similarity between the product search query and the title or the description of the multiple of the plurality of products.

5. A method comprising:

receiving a product search query from a user;

identifying a reference to a color in the product search query;

creating an initial ranking of a plurality of products based on the product search query;

analyzing the product search query to determine whether the user intended an actual color by the reference to the color in the product search query including calculating a cosine similarity between a search query vector and vectors of identified titles to determine whether the user intended the actual color in the product search query;

responsive to determining that the user intended the actual color by the reference to the color in the product search query;

determining one or more additional references of one or more additional colors associated with the actual color in the product search query based on one or more similarity scores each indicating a color distance of a color pair between an individual color of the one or more additional colors and the actual color in the product search query, the one or more similarity scores calculated using a color difference formula;

identifying a plurality of references to the actual color contained in a title or a description of multiple of the plurality of products; and

creating, using one or more processors, an ordered ranking of products of the plurality of products by increasing a ranking of the multiple of the plurality of products that include the reference to the actual color in the title or the description and including in the ordered ranking further products of the plurality of products having the one or more additional references of the one or more additional colors associated with the actual color in the product search query.

6. The method of claim 5, further comprising communicating the ordered ranking of the plurality of products to a device operated by the user.

7. The method of claim 5, further comprising causing to display the ordered ranking of the plurality of products to the user.

8. The method of claim 5, further comprising ranking the plurality of products without using the reference to the color in the product search query in response to a determination that the user did not intend an actual color.

9. The method of claim 5, wherein the analyzing of the product search query to determine whether the user intended the actual color includes calculating the cosine similarity between the product search query and the title or the description of the multiple of the plurality of products.

10. The method of claim 5, wherein the reference to the color in the product search query is a keyword in the product search query.

11. The method of claim 5, wherein the identifying the reference to the color in the product search query includes comparing each term in the product search query with a listing of known color terms.

12. An apparatus comprising:

a memory to store data associated with at least one product; and

one or more processors coupled to the memory, the one or more processors configured to;

receive a product search query from a user;

identify a reference to a color in the product search query;

analyze the product search query to determine whether the user intended an actual color including calculating a cosine similarity between a search query vector and vectors of identified titles to determine whether the user intended an actual color in the product search query;

at least when a determination has been made that the user intended an actual color, determine one or more additional references of one or more additional colors based on one or more similarity scores each indicating a color distance of a color pair between an individual color of the one or more additional colors and the color in the product search query, the one or more similarity scores calculated using a color difference formula;

execute the product search query to identify a plurality of products;

at least when a determination has been made that the user intended an actual color, identify in a description of multiple of the plurality of products one or more references to the color in the product search query;

at least when a determination has been made that the user intended an actual color, create a ranking of the plurality of products by comparing the one or more references to the color in the description of the multiple of the plurality of products to the reference to the color in the product search query and to the one or more additional references of the one or more additional colors; and

at least when a determination has been made that the user did not intend an actual color, create the ranking of the plurality of products without using the reference to the color in the product search query.

13. The apparatus of claim 12, wherein the one or more processors are further configured to communicate the ranking of the plurality of products to a device operated by the user.

14. The apparatus of claim 12, wherein the one or more processors are further configured to identify the reference to the color in the product search query by comparing each term in the product search query with a listing of known color terms.

15. The apparatus of claim 12, wherein a value of 0.8 for the cosine similarity is used as a threshold for determining whether the user intended an actual color in the product search query.