Estimation apparatus, estimation method, program, and integrated circuit

1. An estimation apparatus that estimates an association between content and a user who performed an operation on the content, the apparatus comprising:

a processor;

a receiving unit configured to receive an operation performed on content;

a determining unit configured to determine, using said processor, whether or not an operating user who has performed the operation on the content can be identified based on the operation received by the receiving unit;

a first association information generation unit configured to, when it is determined by the determining unit that the operating user can be identified, generate first association information in which the content subjected to the operation performed by the operating user and the operating user are associated with each other;

a learning unit configured to generate learnt information for associating a given user with content likely to be operated by the given user by using the first association information;

an estimation unit configured to estimate a user who has performed an operation on unidentified content by using the unidentified content and the learnt information generated by the learning unit, the unidentified content being content subjected to the operation when it is determined by the determining unit that the operating user cannot be identified; and

a second association information generation unit configured to generate second association information in which the unidentified content and the user estimated by the estimation unit are associated with each other,

wherein the learning unit is configured to update the learnt information by preferentially using the first association information over the second association information, and

wherein the learning unit is configured to weight the first association information and the second association information such that a weight of the second association information is smaller than a weight of the first association information, and update the learnt information by using the first association information and the second association information that have been weighted.

2. The estimation apparatus according to claim 1 ,

wherein the learning unit is configured to update the learnt information by using only the first association information out of the first association information and the second association information.

3. The estimation apparatus according to claim 1 ,

wherein the learning unit is configured to change the weight of the first association information or the second association information according to a change in the number of pieces of the first association information.

4. The estimation apparatus according to claim 3 ,

wherein the learning unit is configured to:

when the number of pieces of the first association information is not greater than a predetermined threshold value, update the learnt information by using the first association information and the second association information; and

when the number of pieces of the first association information is greater than the threshold value, update the learnt information by using only the first association information.

5. The estimation apparatus according to claim 3 ,

wherein the learning unit is configured to:

when the number of pieces of the first association information is not greater than a predetermined threshold value, update the learnt information by using only the first association information; and

when the number of pieces of the first association information is greater than the threshold value, update the learnt information by using the first association information and the second association information.

6. The estimation apparatus according to claim 1 , further comprising:

a user confirmation unit configured to confirm with a user whether or not the second association information generated by the second association information generation unit is correct; and

an information converting unit configured to convert, into the first association information, the second association information that has been confirmed as correct by the user confirmation unit.

7. An estimation method for estimating an association between content and a user who performed an operation on the content, the method comprising:

receiving an operation performed on content;

determining, using a processor, whether or not an operating user who has performed the operation on the content can be identified based on the operation received in the receiving;

when it is determined in the determining that the operating user can be identified, generating first association information in which the content subjected to the operation performed by the operating user and the operating user are associated with each other;

generating learnt information for associating a given user with content likely to be operated by the given user by using the first association information;

estimating a user who has performed an operation on unidentified content by using the unidentified content and the learnt information generated in the generating of learnt information, the unidentified content being content subjected to the operation when it is determined in the determining that the operating user cannot be identified; and

generating second association information in which the unidentified content and the user estimated in the estimating are associated with each other,

wherein in the generating of learnt information, the learnt information is updated by preferentially using the first association information over the second association information, and

wherein in the generating of learnt information, the first association information and the second association information are weighted such that a weight of the second association information is smaller than a weight of the first association information, and the learnt information is updated by using the first association information and the second association information that have been weighted.

8. A non-transitory computer-readable recording medium on which a program is stored, the program for causing a computer to execute the estimation method according to claim 7 .

9. An integrated circuit provided in an estimation apparatus that estimates an association between content and a user who performed an operation on the content, the integrated circuit comprising:

a processor;

a receiving unit configured to receive an operation performed on content;

a determining unit configured to determine, using said processor, whether or not an operating user who has performed the operation on the content can be identified based on the operation received by the receiving unit;

a first association information generation unit configured to, when it is determined by the determining unit that the operating user can be identified, generate first association information in which the content subjected to the operation performed by the operating user and the operating user are associated with each other;

a learning unit configured to generate learnt information for associating a given user with content likely to be operated by the given user by using the first association information;

an estimation unit configured to estimate a user who has performed an operation on unidentified content by using the unidentified content and the learnt information generated by the learning unit, the unidentified content being content subjected to the operation when it is determined by the determining unit that the operating user cannot be identified; and

a second association information generation unit configured to generate second association information in which the unidentified content and the user estimated by the estimation unit are associated with each other,

wherein the learning unit is configured to update the learnt information by preferentially using the first association information over the second association information, and

wherein the learning unit is configured to weight the first association information and the second association information such that a weight of the second association information is smaller than a weight of the first association information, and update the learnt information by using the first association information and the second association information that have been weighted.