Methods for identifying product variants

Patent metadata

Field	Value
Publication number	US-2025021588-A1
Application number	US-202418904196-A
Country	US
Kind code	A1
Filing date	Oct 2, 2024
Priority date	Apr 20, 2020
Publication date	Jan 16, 2025
Grant date	—

Abstract

A computer-implemented method includes extracting, by one or more processors of one or more computing devices, a product family name from each of a plurality of unstructured product titles associated with a plurality of products. The method further includes determining, by the one or more processors, a degree of similarity between model numbers of the plurality of products. The method further includes determining, by the one or more processors, that at least two of the plurality of products are variants of one another by determining that the at least two of the plurality of products have a same extracted product family name and determining that the degree of similarity between the model numbers of the plurality of products is above a predetermined threshold.

First claim

1 - 20 . (canceled)

21 . A computer-implemented method, comprising:
extracting, by one or more processors of one or more computing devices, a product family name from each of a plurality of unstructured product titles associated with a plurality of products; calculating, by the one or more processors, an edit distance between model numbers of the plurality of products; and determining, by the one or more processors, that at least two of the plurality of products are variants of one another by:
determining that the at least two of the plurality of products have a same extracted product family name; and
determining that the edit distance is above a predetermined threshold, wherein the predetermined threshold is based at least in part on a manufacturer or product type of the plurality of products.

22 . The computer-implemented method of claim 21 , further comprising:
generating, by the one or more processors, a lookup table, the lookup table comprising the variants determined to be in the plurality of products; and storing, by the one or more processors, the lookup table for reference with one or more queries.

23 . The computer-implemented method of claim 22 , further comprising determining, by the one or more processors, a product variant type between the at least two of the plurality of products.

24 . The computer-implemented method of claim 23 , wherein the product variant type is at least one of product color, product finish, product pack size, product size, material type, or weight.

25 . The computer-implemented method of claim 21 , further comprising:
receiving, by the one or more processors, a user query associated with one of the at least two of the plurality of products; and populating, in response to the user query associated with the one of the at least two of the plurality of products, a single webpage of a website with data indicative of the at least two of the plurality of products that are determined to be the variants of one another.

26 . The computer-implemented method of claim 25 , further comprising displaying, by the one or more processors as part of the website, the at least two of the plurality of products that are determined to be variants of one another on the single webpage of the website.

27 . The computer-implemented method of claim 26 , wherein:
the single webpage comprises a first user interface element associated with a first product of the at least two of the plurality of products, the single webpage comprises a second user interface element associated with a second product of the at least two of the plurality of products, and each of the first and second user interface elements are selectable by a user of the single webpage to change a portion of the single webpage for viewing the at least two of the plurality of products that are determined to be variants of one another on the single webpage.

28 . The computer-implemented method of claim 21 , wherein the determining that at least two of the plurality of products are variants of one another further comprises determining that the at least two of the plurality of products have a same brand name.

29 . The computer-implemented method of claim 21 , wherein the determining that at least two of the plurality of products are variants of one another further comprises determining that the at least two of the plurality of products have a same product category.

30 . The computer-implemented method of claim 21 , wherein the edit distance is calculated only for those of the plurality of products that are determined to have the same extracted product family name.

31 . The computer-implemented method of claim 21 , wherein the extracted product family name for a product of the plurality of products represents a portion of an unstructured product title for the product.

32 . A non-transitory computer-readable medium having computer executable instructions stored thereon that, upon execution by a processing device, cause the processing device to perform operations comprising:
extracting a product family name from each of a plurality of unstructured product titles associated with a plurality of products; calculating an edit distance between model numbers of the plurality of products; and determining that at least two of the plurality of products are variants of one another by:
determining that the at least two of the plurality of products have a same extracted product family name; and
determining that the edit distance is above a predetermined threshold, wherein the predetermined threshold is determined based at least in part on a manufacturer or product type of the plurality of products.

33 . The non-transitory computer-readable medium of claim 32 , wherein the instructions further cause the processing device to perform operations comprising determining a product variant type between the at least two of the plurality of products.

34 . The non-transitory computer-readable medium of claim 33 , wherein the instructions further cause the processing device to perform operations comprising determining a product variant property related to the determined product variant type, wherein the product variant property is different for the at least two of the plurality of products that are determined to be variants of one another.

35 . The non-transitory computer-readable medium of claim 34 , wherein the product variant property comprises:
a color of a product when the determined product variant type is product color, a finish of the product when the determined product variant type is product finish, a pack size of the product when the determined product variant type is product pack size, a size of the product when the determined product variant type is product size, a material type of the product when the determined product variant type is material type, or a weight of the product when the determined product variant type is weight.

36 . The non-transitory computer-readable medium of claim 32 , wherein the extracted product family name for a product of the plurality of products represents a portion of an unstructured product title for the product.

37 . The non-transitory computer-readable medium of claim 32 , wherein the plurality of unstructured product titles comprises basic text.

38 . The non-transitory computer-readable medium of claim 32 , wherein the extracting the product family name from each of the plurality of unstructured product titles associated with the plurality of products comprises at least one of:
standardizing terms in the plurality of unstructured product titles with a synonyms dictionary, removing punctuation or other non-standard characters from the plurality of unstructured product titles, or removing numbers or units of measurement from the plurality of unstructured product titles.

39 . The non-transitory computer-readable medium of claim 32 , wherein the extracting the product family name from each of the plurality of unstructured product titles associated with the plurality of products comprises removing words from a manually created blacklist of words.

40 . A computing device, comprising:
a processor configured to perform operations comprising:
extracting, by the processor, a product family name from each of a plurality of unstructured product titles associated with a plurality of products;
calculating, by the processor, an edit distance between model numbers of the plurality of products; and
determining, by the processor, that at least two of the plurality of pr

Assignees

• Home Depot Product Authority Llc

Inventors

• Cui Xiquan
• West Rebecca
• Al Jadda Khalifeh
• Qu Huiming

Classifications

• G06F16/954

  Navigation, e.g. using categorised browsing · CPC title

• G06F16/383

  using metadata automatically derived from the content · CPC title

• G06F16/908

  using metadata automatically derived from the content · CPC title

• G06F16/9035

  Filtering based on additional data, e.g. user or group profiles · CPC title

• G06F16/313

  Selection or weighting of terms for indexing · CPC title