Individualized detailed program recommendations with active updating of viewer preferences
First Claim
1. A method of operating an information processing apparatus including a memory device storing instructions, the method comprising:
- (a) causing a processor to extract a keyword of content data;
(b) causing the processor to select content data among a plurality of content data based on a comparison of the extracted keyword and preference data of a user, wherein the preference data of the user is created through a learning process, a filtering process, and a registration process, wherein the filtering process includes calculating a degree of similarity with a profile of a second user including a similar score to the user for a known feature, the degree of similarity is calculated using a user profile value score for the known feature, a user mean score, a second user profile value score for the known feature, and a second user profile mean;
wherein the degree of similarity is then used, along with the user mean score, the second user profile value score, and the second user profile mean, to calculate a predicted vector, wherein the predicted vector is a numerical value of user interest in an unknown feature to the user,wherein the similar score is created by the registration process and the learning process,wherein the user mean score is a mean value of the user'"'"'s preference data, the second user profile mean represents a mean value of the second user'"'"'s preference data, and the second user profile value score is the second user'"'"'s score for the unknown feature; and
(c) causing the processor to cause a display device to display recommendation information of the selected content data and a plurality of recommended reasons based on the extracted keyword, wherein the recommendation information includes a program name, cast members, and a recommendation level,wherein the plurality of recommended reasons includes the extracted keyword and at least one of an actor and interests of other users, wherein the interests of other users are based on the predicted vector,wherein at least one of the plurality of recommended reasons is changed to an administrator selected word for recommendation that is different than the extracted keyword.
1 Assignment
0 Petitions
Accused Products
Abstract
An information processing apparatus is provided which includes: a first managing part for managing first data about preferences of a user; a second managing part for managing second data about information; a retrieving part for retrieving the second data about the information used by the user; and an updating part for updating the first data by use of the second data retrieved by the retrieving part. If the second data are found to exist within the first data, then the updating part updates the first data using the second data; if the second data are not found to exist within the first data, then the updating part adds the second data to the first data.
-
Citations
28 Claims
-
1. A method of operating an information processing apparatus including a memory device storing instructions, the method comprising:
-
(a) causing a processor to extract a keyword of content data; (b) causing the processor to select content data among a plurality of content data based on a comparison of the extracted keyword and preference data of a user, wherein the preference data of the user is created through a learning process, a filtering process, and a registration process, wherein the filtering process includes calculating a degree of similarity with a profile of a second user including a similar score to the user for a known feature, the degree of similarity is calculated using a user profile value score for the known feature, a user mean score, a second user profile value score for the known feature, and a second user profile mean; wherein the degree of similarity is then used, along with the user mean score, the second user profile value score, and the second user profile mean, to calculate a predicted vector, wherein the predicted vector is a numerical value of user interest in an unknown feature to the user, wherein the similar score is created by the registration process and the learning process, wherein the user mean score is a mean value of the user'"'"'s preference data, the second user profile mean represents a mean value of the second user'"'"'s preference data, and the second user profile value score is the second user'"'"'s score for the unknown feature; and (c) causing the processor to cause a display device to display recommendation information of the selected content data and a plurality of recommended reasons based on the extracted keyword, wherein the recommendation information includes a program name, cast members, and a recommendation level, wherein the plurality of recommended reasons includes the extracted keyword and at least one of an actor and interests of other users, wherein the interests of other users are based on the predicted vector, wherein at least one of the plurality of recommended reasons is changed to an administrator selected word for recommendation that is different than the extracted keyword. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14)
-
-
15. An information processing apparatus comprising:
-
a processor; at least one memory device storing instructions which when executed by the processor, cause the processor to; (a) extract a keyword of content data; (b) select content data among a plurality of content data based on a comparison of the extracted keyword and preference data of a user, wherein the preference data of the user is created through a learning process, a filtering process, and a registration process, wherein the filtering process includes calculating a degree of similarity with a profile of a second user including a similar score to the user for a known feature, the degree of similarity is calculated using a user profile value score for the known feature, a user mean score, a second user profile value score for the known feature, and a second user profile mean; wherein the degree of similarity is then used, along with the user mean score, the second user profile value score, and the second user profile mean, to calculate a predicted vector, wherein the predicted vector is a numerical value of user interest in an unknown feature to the user, wherein the similar score is created by the registration process and the learning process, wherein the user mean score is a mean value of the user'"'"'s preference data, the second user profile mean represents a mean value of the second user'"'"'s preference data, and the second user profile value score is the second user'"'"'s score for the unknown feature; and (c) cause a display device to display recommendation information of the selected content data and a plurality of recommended reasons based on the extracted keyword, wherein the recommendation information includes a program name, cast members, and a recommendation level, wherein the plurality of recommended reasons includes the extracted keyword and at least one of an actor and interests of other users, wherein the interests of other users are based on the predicted vector, wherein at least one of the plurality of recommended reasons is changed to an administrator selected word for recommendation that is different than the extracted keyword. - View Dependent Claims (16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A non-transitory computer-readable medium storing instructions structured to cause an information processing apparatus to:
-
(a) extract a keyword of content data; (b) select content data among a plurality of content data based on a comparison of the extracted keyword and preference data of a user, wherein the preference data of the user is created through a learning process, a filtering process, and a registration process, wherein the filtering process includes calculating a degree of similarity with a profile of a second user including a similar score to the user for a known feature, the degree of similarity is calculated using a user profile value score for the known feature, a user mean score, a second user profile value score for the known feature, and a second user profile mean; wherein the degree of similarity is then used, along with the user mean score, the second user profile value score, and the second user profile mean, to calculate a predicted vector, wherein the predicted vector is a numerical value of user interest in an unknown feature to the user, wherein the similar score is created by the registration process and the learning process, wherein the user mean score is a mean value of the user'"'"'s preference data, the second user profile mean represents a mean value of the second user'"'"'s preference data, and the second user profile value score is the second user'"'"'s score for the unknown feature; and (c) cause a display device to display recommendation information of the selected content data and a plurality of recommended reasons based on the extracted keyword, wherein the recommendation information includes a program name, cast members, and a recommendation level, wherein the plurality of recommended reasons includes the extracted keyword and at least one of an actor and interests of other users, wherein the interests of other users are based on the predicted vector, wherein at least one of the plurality of recommended reasons is changed to an administrator selected word for recommendation that is different than the extracted keyword. - View Dependent Claims (25, 26, 27, 28)
-
Specification