Determining commerce entity pricing and availability based on stylistic heuristics

• US 10,643,258 B2
• Filed: 12/21/2015
• Issued: 05/05/2020
• Est. Priority Date: 12/24/2014
• Status: Active Grant

• Current Assignee: Keep Holdings Liquidating Trust
• Original Assignee: Keep Holdings, Inc.
• Inventors: Kurnit, Scott Philip, Myslinsky, Charles, Quigley, Ryan, Tozzi, Mark, Germaise, Scott Craig
• Primary Examiner(s): Misiaszek, Michael
• Assistant Examiner(s): Warren, Jason B
• Application Number:
  US14/977,330
• Publication Number:
  US 20160189255A1
• Time in Patent Office: 1,597 Days
• CPC Class Codes: G06F 17/00 Digital computing or data processing equipment or methods, specially adapted for specific functions information retrieval, database structures or file system structures therefor G06F16/00
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 30/0283 Price estimation or determination
• CPC Class Codes: G06Q 30/0623 Item investigation
• CPC Class Codes: G06Q 30/0641 Shopping interfaces

Abstract

Disclosed is a system and method for determining pricing and availability of e-commerce entities. The determination includes accessing a web page that includes an entity of interest and generating a visual map representation of the web page that includes elements related to the entity'"'"'s pricing/availability and stylistic identifiers that are applied to those elements. Stylistic identifiers are pricing/availability identifiers based on stylistic heuristics. Certain elements are identified as candidate elements and a likelihood that a candidate element is pricing/availability element of the entity is determined for the candidate elements. A candidate element is selected as the pricing/availability element of the entity by analyzing the determined likelihood of various candidate elements.

Claims

1. A computer-implemented method for determining an entity characteristic, the method comprising:

accessing, by a computer, a web page including information associated with an entity, the entity representing a product or service available for purchase at the web page;

executing, by the computer, the source code associated with the web page to generate one or more elements of the web page;

generate a visual map representation of the one or more elements of the web page, the visual map representation includes stylistic identifiers for the one or more elements, the visual map comprising a coordinate grid that represents each of the one or more elements at a unique position on the coordinate grid;

computing, by the computer, one or more Euclidean vector distances between the one or more elements in the visual map representation using vector algebraic operations on the positions of each of the one or more elements on the coordinate grid of the visual map representation, each Euclidean vector distance indicating a relative magnitude of distance and direction between pairs of the one or more elements in the visual map representation;

identifying one or more candidate elements for the entity characteristic, the one or more candidate elements identified from the one or more elements of the visual map representation;

determining, for each candidate element of the one or more candidate elements, a likelihood that the candidate element is a characteristic element of the entity based at least on the Euclidian vector distances between the candidate element and the other candidate elements of the one or more candidate elements; and

selecting a candidate element of the one or more candidate elements as the characteristic element of the entity by analyzing the determined likelihood of the one or more candidate elements.

2. The computer-implemented method of claim 1, further comprising:

providing information associated with the characteristic element of the entity to a device for presenting the provided information to a user of the device.

3. The computer-implemented method of claim 1, further comprising:

importing information associated with the characteristic element of the entity for storing in a repository associated with a user.

4. The computer-implemented method of claim 1, wherein the entity characteristic is at least one of:

pricing information of the entity and availability status of the entity.

5. The computer-implemented method of claim 4, wherein the one or more candidate elements associated with entity pricing are identified by analyzing text of the one or more elements to identify candidate elements comprising a currency symbol and a number.

6. The computer-implemented method of claim 4, wherein the one or more candidate elements associated with entity availability are identified by analyzing text of the one or more elements to identify candidate elements comprising phrases ‘

add to cart’

or ‘

buy’

.

7. The computer-implemented method of claim 1, wherein the stylistic identifiers comprising at least one of:

price labels, price identifiers, price modifiers, and availability identifiers.

8. The computer-implemented method of claim 1, wherein the likelihood that the candidate element is the entity'"'"'s characteristic element is determined by:

computing a weighted score for the candidate element, the weighted score comprising a plurality of components corresponding to a plurality of stylistic attributes of the candidate element, each of the plurality of stylistic attributes comprising a weighting factor; and

comparing the weighted score with a threshold score.

9. The computer-implemented method of claim 8, wherein the plurality of stylistic attributes of the candidate element comprises at least two or more of:

position within the visual map representation, price value, hierarchical address to retrieve the candidate element, font size, text length, text style, and text content.

10. The computer-implemented method of claim 9, wherein the weighting factor associated with the position within the visual map representation increases the weighted score if the position of the candidate element is above a threshold position within the visual map representation, and decreases the weighted score if the position is below or equal to the threshold position.

11. The computer-implemented method of claim 9, wherein the weighting factor associated with the position within the visual map representation decreases the weighted score if the position of the candidate element has a same coordinate with another candidate element in either a horizontal axis or a vertical axis.

12. The computer-implemented method of claim 9, wherein the weighting factor associated with the price value of the candidate element increases the weighted score if the price value is below a threshold price, and decreases the weighted score if the price value is above or equal to the threshold price, the threshold price is computed by analyzing price information for all candidate elements of the visual map representation.

13. The computer-implemented method of claim 9, wherein the weighting factor associated with the hierarchical address of the candidate element decreases the weighted score if the hierarchical address is similar to a hierarchical address of another candidate element of the visual map representation.

14. The computer-implemented method of claim 9, wherein the weighting factor associated with the font size of the candidate element increases the weighted score if a ratio of the font size of the candidate element to a maximum font size of the visual map representation is above a threshold value, and decreases the weighted score if the ratio is below or equal to the threshold value.

15. The computer-implemented method of claim 9, wherein the weighting factor associated with the text length of the candidate element increases the weighted score if a ratio of the text length of the candidate element to a length of a display container that comprises the text of the candidate element is above a threshold value, and decreases the weighted score if the ratio is below or equal to the threshold value.

16. The computer-implemented method of claim 9, the weighting factor associated with the text style of the candidate element increases the weighted score if the text style comprises bold text, and decreases the weighted score if the text style comprises strikethrough text.

17. The computer-implemented method of claim 9, the weighting factor associated with the text context decreases the weighted score if the text of the candidate element comprises at least one of:

shipping, original, and free.

18. The computer-implemented method of claim 1, wherein determining the likelihood that the candidate element is the characteristic element is implemented by a rules engine comprising one or more rules.

19. The computer-implemented method of claim 1, wherein determining the likelihood that the candidate element is the characteristic element is implemented by a machine learning model that receives the plurality of stylistic attributes of the candidate element and outputs a likelihood that the candidate element is the characteristic element.

20. The computer-implemented method of claim 1, wherein selecting the candidate element as the characteristic element of the entity comprises selecting the candidate element comprising canonical metadata for the entity'"'"'s characteristic information.

21. The computer-implemented method of claim 1, further comprising:

selecting a second candidate element as a characteristic element for an additional entity, the additional entity related to the entity and the second candidate element selected based on the relationship between the entity and the additional entity.

22. The computer-implemented method of claim 21, wherein the additional entity is at least one of:

a useful addition to the entity, a more expensive version of the entity, and a less expensive version of the entity.

23. The computer-implemented method of claim 1, wherein the entity represents a configurable product comprising at least one stylistic attribute of:

size, color, size type, size system, age group, gender, and product category.

24. The computer-implemented method of claim 1, wherein the entity represents at least one of:

a product without physical manifestation, a product downloadable from a network, a product representing a subscription service, and a product representing a digital gift.

25. The computer-implemented method of claim 1, wherein the entity represents a group of products and the selected candidate element represents a bundled price for the group of products.

26. A system for determining an entity characteristic, the system comprising:

a network interface device configured to access a web page including information associated with an entity, the entity representing a product or service available for purchase at the web page;

a processor configured to;

execute the source code associated with the web page to generate one or more elements of the web page;

generate a visual map representation of the one or more elements of the web page, the visual map representation includes stylistic identifiers for the one or more elements, the visual map comprising a coordinate grid that represents each of the one or more elements at a unique position on the coordinate grid;

compute, by the computer, one or more Euclidean vector distances between the one or more elements in the visual map representation using vector algebraic operations on the positions of each of the one or more elements on the coordinate grid of the visual map representation, each Euclidean vector distance indicating a relative magnitude of distance and direction between pairs of the one or more elements in the visual map representation;

identify one or more candidate elements for the entity characteristic, the one or more candidate elements identified from the one or more elements of the visual map representation;

determine, for each candidate element of the one or more candidate elements, a likelihood that the candidate element is a characteristic element of the entity based at least on the Euclidean vector distances between the candidate element and the other candidate elements of the one or more candidate elements; and

select a candidate element of the one or more candidate elements as the characteristic element of the entity by analyzing the determined likelihood of the one or more candidate elements.

27. A non-transitory computer readable storage medium configured to store instructions, the instructions when executed by a processor cause the processor to:

access, by a computer, a web page including information associated with an entity, the entity representing a product or service available for purchase at the web page;

execute, by the computer, the source code associated with the web page to generate one or more elements of the web page;

generate a visual map representation of the one or more elements of the web page, the visual map representation includes stylistic identifiers for the one or more elements, the visual map comprising a coordinate grid that represents each of the one or more elements at a unique position on the coordinate grid;

compute, by the computer, one or more Euclidean vector distances between the one or more elements in the visual map representation using vector algebraic operations on the positions of each of the one or more elements on the coordinate grid of the visual map representation, each Euclidean vector distance indicating a relative magnitude of distance and direction between pairs of the one or more elements in the visual map representation;

identify one or more candidate elements for the entity characteristic, the one or more candidate elements identified from the one or more elements of the visual map representation;

determine, for each candidate element of the one or more candidate elements, a likelihood that the candidate element is a characteristic element of the entity based at least on the Euclidian vector distances between the candidate element and the other candidate elements of the one or more candidate elements; and

select a candidate element of the one or more candidate elements as the characteristic element of the entity by analyzing the determined likelihood of the one or more candidate elements.