Method and apparatus for generating a recommended set of items

• US 10,387,115 B2
• Filed: 09/12/2016
• Issued: 08/20/2019
• Est. Priority Date: 09/28/2015
• Status: Active Grant

• Current Assignee: Direct Cursus Technology LLC
• Original Assignee: Yandex Europe AG (Yandex N.V.)
• Inventors: Lifar, Igor Igorevich, Royzner, Mikhail Aleksandrovich
• Primary Examiner(s): Cao, Phuong Thao
• Application Number:
  US15/262,332
• Publication Number:
  US 20170090867A1
• Time in Patent Office: 1,072 Days
• Field of Search: 707735
• CPC Class Codes: G06F 16/9038 Presentation of query results
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 7/24 Sorting, i.e. extracting data from one or more carriers, rearranging the data in numerical or other ordered sequence, and rerecording the sorted data on the original carrier or on a different carrier or set of carriers sorting methods in generalG06F7/36 takes precedence

Abstract

A method of generating a recommended subset of items for a user of an electronic device, the method being executed at a server, the method comprises: acquiring user events associated with a plurality of users, the user events comprising indications of user queries; for each of the user queries, generating a ranked predicted items list that comprises at least some items from a set of potentially recommendable items, such that each particular item within the ranked predicted items list has an associated rank; for each item within a plurality of ranked predicted items lists, generating, by the server, an item score based on a totality of ranks associated therewith; generating the recommended subset of items from the set of potentially recommendable items by selecting at least one item within the plurality of ranked predicted items lists as the recommended subset of items based on the item scores.

Claims

1. A method of generating a recommended subset of items for a user of an electronic device, the method being executed at a server, the method comprises:

prior to acquiring a request for the recommended subset of items, acquiring, by the server, user events associated with a plurality of users, the user events comprising indications of user queries associated with the plurality of users;

prior to acquiring the request, for each of the user queries, generating, by the server, a ranked predicted items list that comprises at least some items from a set of potentially recommendable items, such that each particular item within the ranked predicted items list has an associated rank;

prior to acquiring the request, for each item within a plurality of ranked predicted items lists, generating, by the server, an item score based on a totality of ranks associated therewith;

prior to acquiring the request, generating, by the server, the recommended subset of items from the set of potentially recommendable items, the generating the recommended subset of items comprises selecting, by the server, at least one item within the plurality of ranked predicted items lists as the recommended subset of items based on the item scores of the items within the plurality of ranked predicted items lists;

acquiring, by the server, the request for the recommended subset of items; and

after acquiring the request, sending a signal, by the server to the respective electronic device, for displaying at least one item selected from the recommended subset of items.

2. The method of claim 1, wherein each ranked predicted items list is associated with a respective user query.

3. The method of claim 2, wherein each indication of the user query comprises a respective user query context.

4. The method of claim 3, the generating the plurality of ranked predicted items lists comprises, for each ranked predicted items list:

inputting, by the server, the respective user query and the user query context into a ranking model algorithm; and

inputting, by the server, the items from the set of potentially recommendable items into the ranking model algorithm.

5. The method of claim 4, wherein the generating the plurality of ranked predicted items lists further comprises, for each ranked predicted items list, retrieving by the server, from the ranking model algorithm a potential ranked predicted items list comprising the items from the set of potentially recommendable items, wherein each item is ranked within the potential ranked predicted items list.

6. The method of claim 5, wherein the generating the plurality of ranked predicted items lists further comprises, for each ranked predicted items list, determining by the server, the ranked predicted items list based on the potential ranked predicted items list, the determining the ranked predicted items list comprises truncating, by the server, the potential ranked predicted items list based on a list threshold, the list threshold being a maximum number of items within the ranked predicted items list.

7. The method of claim 1, wherein the generating the recommended subset of items comprises ranking, by the server, the at least one item within the recommended subset of items based on the respective item scores.

8. The method of claim 7, wherein the ranking the at least one item within the recommended subset of items is further based on the respectively associated ranks of the at least one item.

9. The method of claim 8, wherein the ranking the at least one item within the recommended subset of items is based on the respective item scores and the respectively associated ranks comprises determining, by the server, a respective average associated rank for items within the at least one item having a same item score.

10. A server comprising a hardware processor configured to execute computer-readable instructions and a database for generating a recommended subset of items for a user of an electronic device, the hardware processor being configured to:

prior to acquiring a request for the recommended subset of items, acquire user events associated with a plurality of users, the user events comprising indications of user queries associated with the plurality of users;

prior to acquiring the request, for each of the user queries, generate a ranked predicted items list that comprises at least some items from a set of potentially recommendable items, such that each particular item within the ranked predicted items list has an associated rank;

prior to acquiring the request, for each item within a plurality of ranked predicted items lists, generate an item score based on a totality of ranks associated therewith;

prior to acquiring the request, generate the recommended subset of items from the set of potentially recommendable items, to generate the recommended subset of items the processing module being configured to select at least one item within the plurality of ranked predicted items lists as the recommended subset of items based on the item scores of the items within the plurality of ranked predicted items lists;

acquire the request for the recommended subset of items; and

after acquiring the request, send a signal to the respective electronic device for displaying at least one item selected from the recommended subset of items.

11. The server of claim 10, wherein each ranked predicted items list is associated with a respective user query.

12. The server of claim 11, wherein each indication of the user query comprises a respective user query context.

13. The server of claim 12, to generate the plurality of ranked predicted items lists, the hardware processor is configured to, for each ranked predicted items list:

input the respective user query and the user query context into a ranking model algorithm; and

input the items from the set of potentially recommendable items into the ranking model algorithm.

14. The server of claim 13, wherein to generate the plurality of ranked predicted items lists, the hardware processor is further configured, for each ranked predicted items list, to retrieve from the ranking model algorithm a potential ranked predicted items list comprising the items from the set of potentially recommendable items, wherein each item is ranked within the potential ranked predicted items list.

15. The server of claim 14, wherein to generate the plurality of ranked predicted items lists, the hardware processor is further configured, for each ranked predicted items list, to determine the ranked predicted items list based on the potential ranked predicted items list, to determine the ranked predicted items list the server being configured to truncate the potential ranked predicted items list based on a list threshold, the list threshold being a maximum number of items within the ranked predicted items list.

16. The server of claim 10, wherein to generate the recommended subset of items, the hardware processor is configured to rank the at least one item within the recommended subset of items based on the respective item scores.

17. The server of claim 16, wherein to rank the at least one item within the recommended subset of items, the hardware processor is configured to rank based on the respectively associated ranks of the at least one item.

18. The server of claim 17, wherein the hardware processor is configured to rank the at least one item within the recommended subset of items based on the respective item scores and the respectively associated ranks, the processor being configured to determine a respective average associated rank for items within the at least one item having a same item score.