Method and apparatus for generating a recommended set of items for a user

• US 10,452,731 B2
• Filed: 09/12/2016
• Issued: 10/22/2019
• Est. Priority Date: 09/28/2015
• Status: Active Grant

• Current Assignee: Direct Cursus Technology LLC
• Original Assignee: Yandex Europe AG (Yandex N.V.)
• Inventors: Royzner, Mikhail Aleksandrovich, Lamburt, Victor Grigorievich
• Primary Examiner(s): Gebresenbet, Dinku W
• Application Number:
  US15/262,318
• Publication Number:
  US 20170091336A1
• Time in Patent Office: 1,135 Days
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A method of generating a recommended subset of items for a user of an electronic device is disclosed. The method is executed at a server and comprises identifying a first subset of items within a set of potentially recommendable items based on item features of items within the set of potentially recommendable items. The method also comprises acquiring a request for the recommended subset of items and identifying a second subset of items within the set of potentially recommendable items based on user events associated with the user. Each item within the second subset of items is different from any item within the first subset of items. The method also comprises generating the recommended subset of items which comprises at least some items from the first subset of items and at least some items from the second subset of items.

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:

identifying, by the server, prior to acquiring a request for the recommended subset of items, a first subset of items within a set of potentially recommendable items based on item features of items within the set of potentially recommendable items, the item features being stored in association with the respective items, the identifying being executed, by the server, in an off-line mode, the identifying comprising;

ranking, by the server, the items within the set of potentially recommendable items based on at least one of respective item features associated with a respective item;

the ranking being done based on past interactions of a plurality of users with the items within the set of potentially recommendable items;

the ranking being done without knowledge of a specific user for whom the first subset of items is to be used;

storing, by the server, prior to acquiring the request for the recommended subset of items the first subset of items in a database, the storing being executed, by the server, in the off-line mode;

acquiring, by the server, the request for the recommended subset of items from the electronic device;

identifying, by the server, a second subset of items within the set of potentially recommendable items based on user events associated with the user, the identifying the second subset of items comprising;

identifying, by the server, a user-specific interacted subset of items from the user events associated with the user, the user having interacted with each item within the user-specific interacted subset of items;

identifying, by the server, a potential second subset of items within the set of potentially recommendable items, each item within the potential second subset of items being similar to at least one item within the user-specific interacted subset of items;

excluding, by the server, items from the potential second subset of items that are included within the first subset of items;

the second subset of items comprising at least some items from the potential second subset of items;

generating, by the server, the recommended subset of items by combining at least some items from the first subset of items and at least some items from the second subset of items; and

sending, by the server, instructions for displaying to the user the recommended subset of items.

2. The method of claim 1, the method further comprises ranking, by the server, the items within the recommended subset of items.

3. The method of claim 1, wherein each item within the second subset of items is different from any item within the first subset of items.

4. The method of claim 1, the identifying the potential second subset of items comprises comparing, by the server, each item within the user-specific interacted subset of items with each item within the set of potentially recommendable items.

5. The method of claim 1, the excluding the items from the potential second subset of items that are included within the first subset of items comprises comparing, by the server, each item within the potential second subset of items with each item within the first subset of items.

6. The method of claim 2, wherein the recommended subset of items comprises only a most highly ranked item.

7. The method of claim 2, wherein the recommended subset of items comprises a predetermined number of most highly ranked items.

8. A computer server comprising a non-transitory computer usable information storage medium storing computer-readable instructions and a database, the computer server for generating a recommended subset of items for a user of an electronic device, the computer server, upon executing the computer-readable instructions, being configured to:

identify, prior to acquiring a request for the recommended subset of items, a first subset of items within a set of potentially recommendable items based on item features of items within the set of potentially recommendable items, the item features being stored in association with the respective items, the server being configured to identify the first subset of items in an off-line mode, to identify comprising the server being configured to;

rank the items within the set of potentially recommendable items based on at least one of respective item features associated with a respective item;

the ranking being done based on past interactions of a plurality of users with the items within the set of potentially recommendable items;

the ranking being done without knowledge of a specific user for whom the first subset of items is to be used;

store, prior to acquiring the request for the recommended subset of items, the first subset of items in the database, the server being configured to store the first subset of items in the off-line mode;

acquire the request for the recommended subset of items from the electronic device;

identify a second subset of items within the set of potentially recommendable items based on user events associated with the user, the server configured to identify the second subset of items being configured to;

identify a user-specific interacted subset of items from the user events associated with the user, the user having interacted with each item within the user-specific interacted subset of items;

identify a potential second subset of items within the set of potentially recommendable items, each item within the potential second subset of items being similar to at least one item within the user-specific interacted subset of items;

exclude items from the potential second subset of items that are included within the first subset of items;

the second subset of items comprising at least some items from the potential second subset of items;

generate the recommended subset of items by combining at least some items from the first subset of items and at least some items from the second subset of items; and

send instructions for displaying to the user the recommended subset of items.

9. The server of claim 8, the server being further configured to rank the items within the recommended subset of items.

10. The server of claim 8, wherein each item within the second subset of items is different from any item within the first subset of items.

11. The server of claim 8, to identify the potential second subset of items the server is configured to compare each item within the user-specific interacted subset of items with each item within the set of potentially recommendable items.

12. The server of claim 8, to exclude the items from the potential second subset of items that are included within the first subset of items the server is configured to compare each item within the potential second subset of items with each item within the first subset of items.

13. The server of claim 9, wherein the recommended subset of items comprises only a most highly ranked item.

14. The server of claim 9, wherein the recommended subset of items comprises a predetermined number of most highly ranked items.