Enhancing search queries using user implicit data

• US 10,210,215 B2
• Filed: 04/29/2015
• Issued: 02/19/2019
• Est. Priority Date: 04/29/2015
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Levin, Moshe
• Primary Examiner(s): Allen, William J
• Assistant Examiner(s): Kang, Timothy J
• Application Number:
  US14/699,168
• Publication Number:
  US 20160321727A1
• Time in Patent Office: 1,392 Days
• CPC Class Codes: G06F 16/2457 with adaptation to user needs
• CPC Class Codes: G06F 16/3325 Reformulation based on results of preceding query
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0625 Directed, with specific intent or strategy
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing

Abstract

Systems and methods are disclosed for enhancing user queries. In one embodiment, a method includes receiving user data from a plurality of users in respective user sessions where the user data for each user includes at least one manufacturer-identified term for a product and at least one user term for the product, associating the user term for the product and the manufacturer-identified term for the product in response to receiving the user term and the manufacturer-identified term within a user session from a threshold number of users, and returning search results based on the associated manufacturer-identified term in response to receiving a search query that includes the user term.

Claims

1. A computer-implemented system comprising:

one or more processors; and

one or more computer storage media storing computer-usable instructions that, when used by the one or more processors, cause the system to;

receive user data comprising user interactions with a networked system and from a plurality of users in respective user sessions, the user data for respective user sessions comprising at least one manufacturer-identified term for a product and at least one user term for the product that is not a manufacturer-identified term, each respective user session comprising a period of time where a user searches for the product;

associate the user term for the product with the manufacturer-identified term for the product based on determining the user term and the manufacturer-identified term are each in a number of user sessions, wherein the manufacturer-identified term is determined to be in the number of user sessions based on the number of user sessions including a user action involving the product; and

receive a search query that includes the user term, enhance the search query with the manufacturer-identified term for the product based on the associating of the user term for the product with the manufacturer-identified term for the product, and cause presentation of search results selected using the enhanced search query.

2. The system of claim 1, wherein the user action comprises at least one of selecting the product, purchasing the product, placing the product in a virtual shopping cart, and placing the product on a watch list.

3. The system of claim 1, wherein the user data is received from a remote server.

4. The system of claim 1, wherein the manufacturer-identified term and the user term are received in temporally contiguous submissions from one of the users.

5. The system of claim 1, wherein the user term represents the product in a foreign language.

6. The system of claim 1, wherein the manufacturer-identified term comprises at least one of a brand, a model number, and a part number.

18. The system of claim 1, wherein the system is further caused to determine a score for the association from the number of user sessions, wherein the score is weighted based on a type of the user action in a user session, such that the user action of a first user session of the number of user sessions is given a greater weight in the score than the user action of a second user session of the number of user sessions, the type for the first user session being a purchase action and the type for the second user session being a view product action.

19. The system of claim 1, wherein the system is further caused to determine a score for the association from the number of user sessions, wherein the score is weighted based on a time difference between the user term for the product and the manufacturer-identified term for the product appearing within a user session, such that a first user session of the number of user sessions is given a greater weight in the score than a second user session of the number of user sessions based on the time difference being lower for the first user session than the second user session.

20. The system of claim 1, wherein the associating the user term for the product with the manufacturer-identified term for the product is based on comparing a quantity of the number of user sessions to a threshold value, the threshold value being lowered based on a number of the search results that would be generated without the enhancing of the search query.

7. A computer-implemented method comprising:

receiving user data comprising user interactions with a networked system and from a plurality of users in respective user sessions, the user data for respective user sessions comprising at least one manufacturer-identified term for a product and at least one user term for the product, the user term not included in terms used by a manufacturer to identify the product, each respective user session comprising a period of time where a user searches for the product;

associating the user term for the product with the manufacturer-identified term for the product based on determining the user term and the manufacturer-identified term are each in a number of user sessions, wherein the manufacturer-identified term is determined to be in the number of user sessions based on the number of user sessions including a user action involving the product;

receiving a search query that includes the user term;

enhancing the search query with the manufacturer-identified term for the product based on the associating of the user term for the product with the manufacturer-identified term for the product; and

returning search results using the enhanced search query.

8. The method of claim 7, wherein the user action comprises at least one of selecting the product, purchasing the product, placing the product in a virtual shopping cart, and placing the product on a watch list.

9. The method of claim 7, wherein the user data is received from a remote server.

10. The method of claim 7, wherein the manufacturer-identified term and the user term are received in temporally contiguous submissions from one of the users.

11. The method of claim 7, wherein the user term represents the product in a foreign language.

12. The method of claim 7, wherein the manufacturer-identified term comprises at least one of a brand, a model number, and a part number.

13. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:

receiving user data comprising user interactions with a networked system and from a plurality of users in respective user sessions, the user data for respective user sessions comprising at least one manufacturer-identified term for a product and at least one user term for the product, the user term not included in terms used by a manufacturer to identify the product, each respective user session comprising a period of time where a user searches for the product;

associating the user term for the product with the manufacturer-identified term for the product based on determining the user term and the manufacturer-identified term are each in a number of user sessions, wherein the manufacturer-identified term is determined to be in the number of user sessions based on the number of user sessions including a user action involving the product;

receiving a search query that includes the user term;

enhancing the search query with the manufacturer-identified term for the product based on the associating of the user term for the product with the manufacturer-identified term for the product; and

returning search results using the enhanced search query.

14. The non-transitory machine-readable storage medium of claim 13, wherein the user action comprises at least one of selecting the product, purchasing the product, placing the product in a virtual shopping cart, and placing the product on a watch list.

15. The non-transitory machine-readable storage medium of claim 13, wherein the user data is received from a remote server.

16. The non-transitory machine-readable storage medium of claim 13, wherein the manufacturer-identified term and the user term are received in temporally contiguous submissions from one of the users.

17. The non-transitory machine-readable storage medium of claim 13, wherein the manufacturer-identified term comprises at least one of a brand, a model number, and a part number.