IP Library › Granted Patent US 9,454,581

Granted Patent B1

US 9,454,581 · App. 14/092,575 · Granted Sep 27, 2016

Search with more like this refinements

View Patent ↗

Loading inventors, assignments & file history…

Monitor This Case

Get email alerts when status or documents change.

Order Certified Copies

Most orders are placed with the USPTO same day — all within 24 business hours.

Order via The Patent Place →

Pre-filled with this patent's details

Quick Facts

Patent No.: US 9,454,581
App. No.: 14/092,575
Granted: Sep 27, 2016
Kind: B1

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

Techniques for search with more like this refinements are disclosed. In some embodiments, search with more like this refinements includes receiving a product and a context (e.g., the context can include related category information, user context, and/or other context related information); generating a search query based on the product and the context; and determining a plurality of products that match the search query to generate more like this search results.

Claims (101)

1. A system for search with more like this refinements, comprising:

a processor configured to:

receive a product and a context;

determine a product class, comprising to:

weigh a first product by a first weight to obtain a first weighted product, the first weight relating to a strength of association of the first product to the product class;

weigh a second product by a second weight to obtain a second weighted product, the second weight relating to a strength of association of the second product to the product class; and

determine the product class based at least in part on the first and second weighted products and a function;

generate a search query based on the product, the product class, and the context;

determine a field boost associated with the search query, comprising to:

determine a first match of a first string of the search query with a first field of a merchant's web page including the product;

determine a second match of a second string of the search query with a second field of the merchant's web page including the product, the first field being different from the second field;

weigh the first field with a third weight to obtain a first weighted field;

weigh the second field with a fourth weight to obtain a second weighted field, the third weight being different from the fourth weight; and

obtain the field boost based at least in part on the first and second weighted fields; and

determine a plurality of products that match the search query to generate more like this search results, wherein the plurality of products that match the search query is determined based at least in part on the field boost; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system recited in claim 1 , wherein the memory is further configured to:

receive a product identifier to uniquely identify the product.

3. The system recited in claim 1 , wherein the memory is further configured to:

determine a user context.

4. The system recited in claim 1 , wherein the memory is further configured to:

determine a user context, wherein the user context includes user profile information.

5. The system recited in claim 1 , wherein the memory is further configured to:

determine a user context, wherein the user context includes user browsing history information.

6. The system recited in claim 1 , wherein the memory is further configured to:

determine a user context, wherein the search query is generated based on the user context.

7. The system recited in claim 1 , wherein the memory is further configured to:

determine a product attribute associated with the product.

8. The system recited in claim 1 , wherein the memory is further configured to:

determine a product attribute associated with the product, wherein the product attribute is a merchant specified product attribute.

9. The system recited in claim 1 , wherein the memory is further configured to:

determine a product attribute associated with the product, wherein the product attribute is a derived product attribute.

10. The system recited in claim 1 , wherein the memory is further configured to:

determine a product attribute associated with the product, wherein the search query is generated based on the product attribute.

11. The system recited in claim 1 , wherein the function relates to a decaying algorithm.

12. A method of search with more like this refinements, comprising:

receiving a product and a context;

determining a product class, comprising:

weighing a first product by a first weight to obtain a first weighted product, the first weight relating to a strength of association of the first product to the product class;

weighing a second product by a second weight to obtain a second weighted product, the second weight relating to a strength of association of the second product to the product class; and

determining the product class based at least in part on the first and second weighted products and a function;

generating a search query based on the product, the product class, and the context;

determining a field boost associated with the search query, comprising to:

determining a first match of a first string of the search query with a first field of a merchant's web page including the product;

determining a second match of a second string of the search query with a second field of the merchant's web page including the product, the first field being different from the second field;

weighing the first field with a third weight to obtain a first weighted field;

weighing the second field with a fourth weight to obtain a second weighted field, the third weight being different from the fourth weight; and

obtaining the field boost based at least in part on the first and second weighted fields; and

determining a plurality of products that match the search query to generate more like this search results, wherein the plurality of products that match the search query is determined based at least in part on the field boost.

13. The method of claim 12 , further comprising:

receiving a product identifier to uniquely identify the product.

14. The method of claim 12 , further comprising:

determining a user context.

15. The method of claim 12 , further comprising:

determining a user context, wherein the user context includes user profile information.

16. The method of claim 12 , further comprising:

determining a user context, wherein the user context includes user browsing history information.

17. The method of claim 12 , further comprising:

determining a user context, wherein the search query is generated based on the user context.

18. The method of claim 12 , further comprising:

determining a product attribute associated with the product.

19. The method of claim 12 , further comprising:

determining a product attribute associated with the product, wherein the product attribute is a merchant specified product attribute.

20. The method of claim 12 , further comprising:

determining a product attribute associated with the product, wherein the product attribute is a derived product attribute.

21. The method of claim 12 , further comprising:

determining a product attribute associated with the product, wherein the search query is generated based on the product attribute.

22. The method of claim 12 , wherein the function relates to a decaying algorithm.

23. A computer program product for search with more like this refinements, the computer program product comprising a tangible non-transitory computer readable storage medium and comprising computer instructions for: