Content recommendation system, content recommendation device, and content recommendation method

• US 9,536,246 B2
• Filed: 05/26/2011
• Issued: 01/03/2017
• Est. Priority Date: 06/03/2010
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Saito, Katsu, Onogi, Wataru, Koyanagi, Masaaki, Sakamoto, Takayuki
• Primary Examiner(s): Lai, Andrew
• Assistant Examiner(s): Rahman, M Mostazir
• Application Number:
  US13/116,932
• Publication Number:
  US 20110302240A1
• Time in Patent Office: 2,049 Days
• Field of Search: 709/203
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9035 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
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04L 65/612 for unicast

Abstract

A content recommendation system includes: a client terminal operated by a user; and a content recommendation server that recommends content to the terminal through a network and includes a preference discrimination information storage section accumulating and storing preference discrimination information in which information related to evaluation of content input to the terminal by a user and identification information of the user are associated with each other, a content feature information storage section storing content feature information in which data indicating features of content are associated with identification information of content, a preference information creation section creating preference information of each user by integrating evaluation for content with the feature of the content for each user based on the preference discrimination information and the content feature information, and a recommended content list providing section transmitting a list of recommended content, extracted by evaluating the degree of similarity of features with the preference information based on the content feature information, to the terminal.

Claims

1. A content recommendation system, comprising:

a client terminal operated by a user; and

a recommendation server comprising a first one or more processors configured to;

recommend content to the client terminal through a network;

accumulate and store preference discrimination information in which information related to an evaluation for content input to the client terminal by the user and identification information of the user are associated with each other;

store content feature information in which data that indicates features of the input content is associated with identification information of the input content;

create preference information of the user by integration of the evaluation for the input content with the content feature information, wherein the preference information is created based on an evaluation time at which the user performs the evaluation for the input content at the client terminal;

transmit, to the client terminal, a list of the recommended content, extracted by integration of a degree of similarity of the content feature information with the preference information,

wherein the preference information includes an evaluation date;

wherein a second one or more processors in the client terminal are configured to;

receive an input to designate an evaluation time slot as a condition for the creation of the preference information from the user operating the client terminal; and

transmit the input to designate the evaluation time slot to the recommendation server; and

wherein the first one or more processors in the recommendation server are configured to create preference information to limit a time slot based on the content feature information and the preference discrimination information evaluated at the evaluation time slot designated by the user.

2. The content recommendation system according to claim 1, wherein the second one or more processors are further configured to receive an input to designate conditions of extraction for a first user from a second user that operates the client terminal, and transmit the input to the recommendation server.

3. The content recommendation system according to claim 2, wherein the first one or more processors are further configured to create, preference information for a user set including a plurality of users extracted using the conditions transmitted from the client terminal, by integration of preference information of the plurality of users.

4. The content recommendation system according to claim 1, wherein the first one or more processors are further configured to create preference information of a user set by integration of preference information of users that belongs to the user set clustered in accordance with a predetermined rule.

5. The content recommendation system according to claim 1, wherein the first one or more processors are configured to:

store the content feature information of a plurality of types of content having different reproduction formats, and

set a type of content to the recommended content, different from a type of content evaluated by the user that operates the client terminal.

6. The content recommendation system according to claim 1, wherein the first one or more processors are configured to:

transmit the created preference information to the client terminal,

wherein the second one or more processors in the client terminal are further configured to;

generate request data obtained by addition of filtering conditions to the preference information acquired from the recommendation server, and

transmit the request data to the recommendation server, and

wherein the first one or more processors are configured to filter using the filtering conditions designated by the request data at the time of extraction of the list of recommended content.

7. The content recommendation system according to claim 6, wherein the first one or more processors are configured to transmit the preference information in a format which makes the content unreadable by the user of the client terminal.

8. The content recommendation system according to claim 1, wherein the information related to the evaluation for the input content included in the preference discrimination information includes an evaluated place,

wherein the second one or more processors in the client terminal are configured to:

receive an input that designates an evaluated area as a condition for creation of preference information from the user operating the client terminal, and

transmit the input that designates the evaluated area to the recommendation server, and

wherein the first one or more processors preference information creation section in the recommendation server are configured to create preference information to limit an evaluation area based on the content feature information and the preference discrimination information evaluated within the evaluated area designated by the user.

9. The content recommendation system according to claim 1, wherein the first one or more processors in the recommendation server are configured to:

compare combinations of content to be evaluated by same user with reference to the preference discrimination information,

evaluate a level of relevance between content items of the combination by a plurality of users that evaluate similar combination, and in an event the degree of similarity of the content feature information with the preference information is evaluated, perform weighting on the degree of similarity in accordance with the level of relevance.

10. The content recommendation system according to claim 1, wherein after lapse of a particular time period from a time at which execution frequency of the content falls below a particular execution frequency, a new content is recommended to the user of the client terminal.

11. A content recommendation device, connected to a plurality of client terminals through a network, that recommends content to each of the plurality client terminals, the content recommendation device comprising:

a first one or more processors configured to;

recommend the content to a client terminal of the plurality of client terminals through the network;

accumulate and store preference discrimination information in which information related to an evaluation for content input to a client terminal of the plurality of client terminals by a user and identification information of the user are associated with each other;

store content feature information in which data that indicates features of the input content is associated with identification information of the input content;

create preference information of the user by integration of the evaluation for the input content with the content feature information, wherein the preference information is created based on an evaluation time at which the user performs the evaluation for the input content at the client terminal;

transmit, to the client terminal, a list of the recommended content, extracted by evaluation of a degree of similarity of the content feature information with the preference information,

wherein the preference information includes an evaluation date;

wherein a second one or more processors in the client terminal are configured to;

receive an input for designation of an evaluation time slot as a condition for the creation of the preference information from the user operating the client terminal; and

transmit the input for designation of the evaluation time slot to the recommendation server; and

wherein the first one or more processors in the recommendation server are configured to create preference information to limit a time slot based on the content feature information and the preference discrimination information evaluated at the evaluation time slot designated by the user.

12. The content recommendation device according to claim 11, wherein the one or more processors are configured to:

compare combinations of content to be evaluated by same user with reference to the preference discrimination information;

evaluate a level of relevance between content items of the combination by a plurality of users evaluate similar combination, wherein the preference discrimination information is created based on an evaluation time at which the user performs the evaluation for the content at the client terminal;

create a list of recommended content by reflection of the evaluation result to extraction processing of recommend content, and transmit the list of recommended content to the client terminal of the plurality of client terminals; and

set a type of content to the recommended content, different from a type of content evaluated by the user that operate the client terminal.

13. A content recommendation method, comprising:

in a recommendation server communicatively coupled to a client terminal operated by a user, the recommendation server comprising a memory, and a first one or more processors;

recommending, by the first one or more processors, content to the client terminal through a network;

accumulating and storing in the memory, by the first one or more processors, preference discrimination information, in which information related to an evaluation for content input to the client terminal by the user and identification information of the user are associated with each other;

storing in the memory, by the first one or more processors, content feature information, in which data indicating features of the input content is associated with identification information of the input content;

creating by the first one or more processors, preference information of the user by integrating the evaluation for the input content with the content feature information for the user, wherein the preference information is created based on an evaluation time at which the user performs the evaluation for the input content at the client terminal; and

transmitting, by the first one or more processors, to the client terminal, a list of the recommended content, extracted by evaluating a degree of similarity of the content feature information with the preference information,

wherein the preference information includes an evaluation date;

wherein a second one or more processors in the client terminal are configured to;

receive an input for designating an evaluation time slot as a condition for the creation of the preference information from the user operating the client terminal; and

transmit the input for designating the evaluation time slot to the recommendation sever; and

wherein the first one or more processors in the recommendation server are configured to create preference information to limit a time slot based on the content feature information and the preference discrimination information evaluated at the evaluation time slot designated by the user.

14. The method of claim 13, further comprising:

transmitting a list of recommended content to the client terminal operated by the user, wherein the recommended content is based on preference information of another user; and

wherein the client terminal is configured to receive an input for designating conditions of extraction for the user from the other user operating another client terminal, and transmit the input to the recommendation server.

15. The method of claim 13, further comprising:

creating, preference information of a user set including a plurality of users extracted using conditions transmitted from the client terminal, by integrating preference information of the plurality of users.

16. The method of claim 13, further comprising:

creating preference information of each user set by integrating preference information of users belonging to a user set clustered in accordance with a predetermined rule.

17. A non-transitory computer-readable medium, storing thereon computer executable instructions that cause a first one or more processors in a recommendation server to perform operations, comprising:

recommending content to a client terminal operated by a user through a network;

accumulating and storing in a memory of the recommendation server, by the first one or more processors in the recommendation server, preference discrimination information in which information related to an evaluation for content input to the client terminal by the user and identification information of the user are associated with each other;

storing in the memory, by the first one or more processors in the recommendation server, content feature information, in which data indicating features of the input content is associated with identification information of the input content;

creating by the first one or more processors, preference information for the user by integrating the evaluation for the input content with the content feature information for the user, wherein the preference information is created based on an evaluation time at which the user performs the evaluation for the input content at the client terminal; and

transmitting, by the first one or more processors, to the client terminal, a list of recommended content, extracted by evaluating a degree of similarity of the content feature information with the preference information,

wherein the preference information includes an evaluation date;

wherein a second one or more processors in the client terminal are configured to;

receive an input for designating an evaluation time slot as a condition for the creation of the preference information from the user operating the client terminal; and

transmit the input for designating the evaluation time slot to the recommendation sever; and

wherein the first one or more processors in the recommendation server are configured to create preference information to limit a time slot based on the content feature information and the preference discrimination information evaluated at the evaluation time slot designated by the user.

18. The non-transitory computer-readable medium of claim 17, wherein the computer executable instructions further comprises code that cause the first one or more processors to perform:

transmitting a list of recommended content to the client terminal operated by the user, wherein the recommended content is based on preference information of another user,

wherein the client terminal is configured to receive an input for designating conditions of extraction for the user from the other user operating another client terminal, and transmit the input to the recommendation server.

19. The non-transitory computer-readable medium of claim 17, wherein the computer executable instructions further comprises code that cause the first one or more processors to perform:

creating, preference information of a user set including a plurality of users extracted by the conditions transmitted from the other client terminal, by integrating preference information of the plurality of users.

20. The non-transitory computer-readable medium of claim 17, wherein the computer executable instructions further comprises code that cause the first one or more processors to perform:

transmitting a list of recommended content, based on preference information of a user different from the user operating the client terminal which is a destination of the list of recommended content, to the client terminal, and

wherein the client terminal is configured to receive an input for designating conditions of extraction for the user different from the user operating the client terminal, and transmit the input to the recommendation server.