Information processing apparatus, information processing method, and program for prediction model generated based on evaluation information

• US 9,785,888 B2
• Filed: 11/13/2015
• Issued: 10/10/2017
• Est. Priority Date: 05/31/2010
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Tateno, Kei
• Primary Examiner(s): CHEN, ALAN S
• Application Number:
  US14/940,834
• Publication Number:
  US 20160071015A1
• Time in Patent Office: 697 Days
• Field of Search: None
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: H04N 21/4661 Deriving a combined profile for a plurality of end-users of the same client, e.g. for family members within a home user profiles in network data switching protocols H04L67/306

Abstract

Disclosed herein is an information processing apparatus including an evaluation information extraction section configured to extract evaluation information including an object targeted to be evaluated and an evaluation of the object targeted to be evaluated from a linguistic expression given as information expressed linguistically by a user of interest; an identification section configured to identify whether the evaluation information is of a first type regarding content or of a second type regarding another user; and an evaluation prediction section configured to predict the evaluation by the user of interest regarding the content, based on the evaluation information of the first type given by the user of interest and on the evaluation information given by the other user in the evaluation information of the second type given by the user of interest.

Claims

1. An information processing server comprising:

circuitry configured to;

receive first evaluation information given by a first user, the first evaluation information including content type evaluation and person type evaluation;

analyze the received first evaluation information, including determining whether the first evaluation information includes positive evaluation to a second user or positive evaluation to a content,

wherein,

if a result of the analysis indicates that the first evaluation information includes the positive evaluation to the second user, the circuitry is further configured to generate a prediction model of the first user based on the first evaluation information of the first user and second evaluation information of the second user evaluated positively by the first user, and

if the result of the analysis indicates that the first evaluation information does not include the positive evaluation to the second user, but includes the positive evaluation to the content, the circuitry is further configured to generate the prediction model of the first user based on the first evaluation information of the first user; and

output a recommended content to a user terminal associated with the first user based on the prediction model of the first user.

2. The information processing server of claim 1, wherein the circuitry is further configured to generate a set of parameters corresponding to a set of content features as the prediction model of the first user.

3. The information processing server of claim 1, wherein the circuitry is further configured to generate the prediction model of the first user by incorporating a prediction model of the second user when the result of the analysis indicates that the second user is evaluated positively by the first user.

4. The information processing server of claim 1, wherein the circuitry is further configured to generate the prediction model of the first user by merging a prediction model of the second user with a stored prediction model of the first user in a weighted-average manner when the result of the analysis indicates that the second user is evaluated positively by the first user.

5. The information processing server of claim 1, wherein the circuitry is further configured to:

generate a base prediction model of the first user based on evaluation given by the first user regarding evaluated contents; and

modify the base prediction model to generate the prediction model of the first user based on the evaluation given by the first user regarding the second user.

6. The information processing server of claim 1, wherein the circuitry is further configured to extract the first evaluation information from linguistic expression input by the first user on a web page.

7. The information processing server of claim 1, wherein the circuitry is further configured to output the recommended content based on a set of content feature values of the recommended content and the prediction model.

21. The information processing server of claim 1, wherein the circuitry is further configured to:

generate the prediction model of the first user based on reversing favorability included in another evaluation information given by the second user when the result of the analysis indicates that the second user is evaluated negatively by the first user.

8. An information processing method comprising:

receiving first evaluation information given by a first user, the first evaluation information including content type evaluation and person type evaluation;

analyzing the received first evaluation information, including determining whether the first evaluation information includes positive evaluation to a second user or positive evaluation to a content,

wherein,

if a result of the analysis indicates that the first evaluation information includes the positive evaluation to the second user, the method further includes generating a prediction model of the first user based on the first evaluation information of the first user and second evaluation information of the second user evaluated positively by the first user, and

if the result of the analysis indicates that the first evaluation information does not include the positive evaluation to the second user, but includes the positive evaluation to the content, the method further includes generating the prediction model of the first user based on the first evaluation information of the first user; and

outputting a recommended content to a user terminal associated with the first user based on the prediction model of the first user.

9. The method of claim 8, wherein the generating the prediction model of the first user comprises generating a set of parameters corresponding to a set of content features as the prediction model of the first user.

10. The method of claim 8, wherein the generating the prediction model of the first user comprises incorporating a prediction model of the second user when the result of the analysis indicates that the second user is evaluated positively by the first user.

11. The method of claim 8, wherein the generating the prediction model of the first user comprises merging a prediction model of the second user with a stored prediction model of the first user in a weighted-average manner when the result of the analysis indicates that the second user is evaluated positively by the first user.

12. The method of claim 8, wherein the generating the prediction model of the first user comprises:

generating a base prediction model of the first user based on evaluation given by the first user regarding evaluated contents; and

modifying the base prediction model to generate the prediction model of the first user based on the evaluation given by the first user regarding the second user.

13. The method of claim 8, further comprising extracting the first evaluation information from linguistic expression input by the first user on a web page.

14. The method of claim 8, further comprising outputting the recommended content based on a set of content feature values of the recommended content and the prediction model.

15. A non-transitory, computer-readable medium comprising instructions for causing a computer to execute a process comprising:

receiving first evaluation information given by a first user, the first evaluation information including content type evaluation and person type evaluation;

analyzing the received first evaluation information, including determining whether the first evaluation information includes positive evaluation to a second user or positive evaluation to a content,

wherein,

if a result of the analysis indicates that the first evaluation information includes the positive evaluation to the second user, the process further includes generating a prediction model of the first user based on the first evaluation information of the first user and second evaluation information of the second user evaluated positively by the first user, and

if the result of the analysis indicates that the first evaluation information does not include the positive evaluation to the second user, but includes the positive evaluation to the content, the process further includes generating the prediction model of the first user based on the first evaluation information of the first user; and

outputting a recommended content to a user terminal associated with the first user based on the prediction model of the first user.

16. The non-transitory, computer-readable medium of claim 15, wherein the generating the prediction model of the first user comprises generating a set of parameters corresponding to a set of content features as the prediction model of the first user.

17. The non-transitory, computer-readable medium of claim 15, wherein the generating the prediction model of the first user comprises incorporating a prediction model of the second user when the result of the analysis indicates that the second user is evaluated positively by the first user.

18. The non-transitory, computer-readable medium of claim 15, wherein the generating the prediction model of the first user comprises merging a prediction model of the second user with a stored prediction model of the first user in a weighted-average manner when the result of the analysis indicates that the second user is evaluated positively by the first user.

19. The non-transitory, computer-readable medium of claim 15, wherein the generating the prediction model of the first user comprises:

generating a base prediction model of the first user based on evaluation given by the first user regarding evaluated contents; and

modifying the base prediction model to generate the prediction model of the first user based on the evaluation given by the first user regarding the second user.

20. The non-transitory, computer-readable medium of claim 15, wherein the process further comprises extracting the first evaluation information from linguistic expression input by the first user on a web page.