Recommendation for diverse content

• US 9,946,761 B2
• Filed: 08/13/2014
• Issued: 04/17/2018
• Est. Priority Date: 07/14/2009
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Saito, Katsu, Isozaki, Masaaki, Onogi, Wataru, Ishii, Kazuo, Ikeda, Nozomu, Takahashi, Yoshikazu
• Primary Examiner(s): Fan, Shiow-Jy
• Application Number:
  US14/459,206
• Publication Number:
  US 20140351207A1
• Time in Patent Office: 1,343 Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/4387 by the use of playlists
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

The content recommendation system that acquires attribute information of a given user, acquires at least one list from among lists of content sequentially generated over time on the basis of the attribute information of the given user, acquires user preference information, which is feature information of content preferred by the given user, extracts some content from content included in the list acquired on the basis of the user preference information, and presents the content to the given user.

Claims

1. An apparatus, comprising:

a memory; and

a processor that;

recommends a first number of times a plurality of content items to a user,

receives, from a user device of the user, a second number of times positive evaluation information for the plurality of content items,

wherein the positive evaluation information indicates that a content item of the plurality of content items is completely reproduced by the user device, and

wherein the plurality of content items are associated with a plurality of content item IDs;

generates rankings of the plurality of content items, wherein the rankings of the plurality of content items are indicated by a first list of the plurality of content item IDs in a descending order of a ratio of the first number of times the positive evaluation information is received from the user device for the content item and the second number of times the content item is recommended to the user;

extracts at least two content item IDs from the first list of the plurality of content item IDs,

wherein the extraction is based on the rankings of the plurality of content items and on user preference information associated with the content item, and

wherein the user preference information comprises a user preference vector;

sorts the extracted at least two content item IDs in a second list in an increasing order of an angle between the user preference vector and a plurality of feature vectors associated with the extracted at least two content item IDs,

wherein each of the plurality of feature vectors comprises at least a feature quantity associated with the at least two content item IDs, and

wherein the user preference vector is generated based on the positive evaluation information; and

transmits the second list to the user device.

2. The apparatus of claim 1, wherein the processor further:

receives a plurality of first attributes of the user and browsing information of the user, and

generates aggregated browsing information of the user based on combinations of the plurality of content item IDs and a plurality of categories associated with the plurality of first attributes of the user; and

generates rankings of subsets of the plurality of content items,

wherein the subsets correspond to the plurality of categories, and

wherein the generation of the rankings of the subsets is based on changes in the aggregated browsing information.

3. The apparatus of claim 2, wherein the processor further:

receives a plurality of second attributes of the user, and

generates a synthesized list of the plurality of content item IDs based on the rankings of the subsets, wherein the subsets correspond to the plurality of categories encompassing the plurality of second attributes,

filters the synthesized list of the plurality of content item IDs based on the user preference information associated with the content item,

sorts the filtered synthesized list of the plurality of content item IDs based on the user preference information associated with the content item, and

transmits the sorted filtered synthesized list of the plurality of content item IDs to the user device.

4. The apparatus of claim 3, wherein the user preference information associated with the content item comprises first content attribute information.

5. The apparatus of claim 4, wherein the processor further receives the user preference information associated with the content item from the user device, based on a user designation of the first content attribute information.

6. The apparatus of claim 4, wherein the filtration comprises:

a reception of second content attribute information for the plurality of content item IDs in the synthesized list from a content attribute database, and

a comparison of the received second content attribute information with the first content attribute information of the user preference information associated with the content item.

7. The apparatus of claim 3, wherein the user preference vector is updated based on the positive evaluation information.

8. The apparatus of claim 3, wherein values of elements of the plurality of feature vectors are based on the plurality of content item IDs.

9. The apparatus of claim 8, wherein the user preference vector increments towards a feature vector, of the plurality of feature vectors, for a content item ID of the plurality of content item IDs, and

wherein the increment is based on the positive evaluation information of the content item associated with the content item ID.

10. The apparatus of claim 3, wherein a first content distributor of the sorted filtered synthesized list is different from a second content distributor of the content item.

11. The apparatus of claim 2,

wherein the browsing information of the user includes indications of a user access to a text, and

wherein the text is related to the plurality of content items.

12. The apparatus of claim 11, wherein the changes in the aggregated browsing information is based on an increase in a number of the indications of the user access for a combination of the combinations of the plurality of content item IDs and a category of the plurality of categories.

13. The apparatus of claim 2,

wherein the plurality of first attributes include at least one of an age, a residence, or a language, and

wherein the plurality of categories include at least one of an age range, the residence, or the language.

14. A method, comprising:

recommending a first number of times a plurality of content items to a user,

receiving, from a user device of the user, a second number of times positive evaluation information for the plurality of content items,

wherein the positive evaluation information indicates that a content item of the plurality of content items is completely reproduced by the user device, and

wherein the plurality of content items are associated with a plurality of content item IDs;

generating rankings of the plurality of content items, wherein the rankings of the plurality of content items are indicated by a first list of the plurality of content item IDs in a descending order of a ratio of the first number of times the positive evaluation information is received from the user device for the content item and the second number of times the content item is recommended to the user;

extracting at least two content item IDs from the first list of the plurality of content item IDs,

wherein the extraction is based on the rankings of the plurality of content items and a user preference information associated with the content item, and

wherein the user preference information comprises a user preference vector;

sorting the extracted at least two content item IDs in a second list in an increasing order of an angle between the user preference vector and a plurality of feature vectors associated with the extracted at least two content item IDs,

wherein each of the plurality of feature vectors comprises at least a feature quantity associated with the at least two content item IDs, and

wherein the user preference vector is generated based on the positive evaluation information; and

transmitting the second list to the user device.

15. The method of claim 14, further comprising:

receiving a plurality of first attributes of the user and browsing information of the user;

aggregating the browsing information of the user based on combinations of the plurality of content item IDs and a plurality of categories associated with the plurality of first attributes of the user; and

generating rankings of subsets of the plurality of content items,

wherein the subsets corresponding to the plurality of categories, and

wherein the generation of the rankings of the subsets is based on changes in the aggregated browsing information.

16. The method according to claim 15, further comprising:

receiving a plurality of second attributes of the user and the user preference information associated with the content item;

generating a synthesized list of the plurality of content item IDs based on the rankings of the subsets, the plurality of second attributes of the user, and the user preference information associated with the content item;

filtering the synthesized list of the plurality of content item IDs based on the user preference information associated with the content item;

sorting the filtered synthesized list of the plurality of content item IDs based on the user preference information associated with the content item; and

transmitting the sorted filtered synthesized list of the plurality of content item IDs to the user device.

17. The method of claim 16, wherein the user preference information associated with the content item comprises content attribute information.

18. The method of claim 17, further comprising receiving the user preference information associated with the content item from the user device, based on a user designation of the content attribute information.

19. The method of claim 15, wherein the browsing information includes indications of a user access to a text, and wherein the text is related to the plurality of content items.

20. The method of claim 14, further comprising updating the user preference vector based on the positive evaluation information.

21. A non-transitory computer-readable medium having stored thereon computer-executable instructions that, when executed by a processor, cause a computer to perform operations, the operations comprising:

recommending a first number of times a plurality of content items to a user,

receiving, from a user device of the user, a second number of times positive evaluation information for the plurality of content items,

wherein the positive evaluation information indicates that a content item of the plurality of content items is completely reproduced by the user device, and

wherein the plurality of content items are associated with a plurality of content item IDs;

generating rankings of the plurality of content items, wherein the rankings of the plurality of content items are indicated by a first list of the plurality of content item IDs in a descending order of a ratio of the first number of times the positive evaluation information is received from the user device for the content item and the second number of times the content item is recommended to the user;

extracting at least two content item IDs from the first list of the plurality of content item IDs,

wherein the extraction is based on the rankings of the plurality of content items and a user preference information associated with the content item, and

wherein the user preference information comprises a user preference vector;

sorting the extracted at least two content item IDs in a second list in an increasing order of an angle between the user preference vector and a plurality of feature vectors associated with the extracted at least two content item IDs,

wherein each of the plurality of feature vectors comprises at least a feature quantity associated with the at least two content item IDs, and

wherein the user preference vector is generated based on the positive evaluation information; and

transmitting the second list to the user device.

22. The non-transitory computer-readable medium of claim 21, further comprising:

receiving a plurality of first attributes of the user and browsing information of the user;

aggregating the browsing information of the user based on combinations of the plurality of content item IDs and a plurality of categories associated with the plurality of first attributes of the user; and

generating rankings of subsets of the plurality of content items,

wherein the subsets corresponding to the plurality of categories, and

wherein the generation of the rankings of the subsets is based on changes in the aggregated browsing information.

23. The non-transitory computer-readable medium of claim 22, further comprising:

receiving a plurality of second attributes of the user and the user preference information associated with the content item;

generating a synthesized list of the plurality of content item IDs based on the rankings of the subsets, the plurality of second attributes of the user, and the user preference information associated with the content item;

filtering the synthesized list of the plurality of content item IDs based on the user preference information associated with the content item;

sorting the filtered synthesized list of the plurality of content item IDs based on the user preference information associated with the content item; and

transmitting the sorted filtered synthesized list of the plurality of content item IDs to the user device.

24. The non-transitory computer-readable medium of claim 23, wherein the user preference information associated with the content item comprises content attribute information.

25. The non-transitory computer-readable medium of claim 24, further comprising receiving the user preference information associated with the content item from the user device, based on a user designation of the content attribute information.

26. The non-transitory computer-readable medium of claim 22, wherein the browsing information includes indications of a user access to a text, and wherein the text is related to the plurality of content items.

27. The non-transitory computer-readable medium of claim 21, further comprising updating the user preference vector based on the positive evaluation information.