Information-processing apparatus, method, system, computer-readable medium and method for automatically recording or recommending content
First Claim
1. An information processing apparatus, comprising:
- circuitry configured to generate a content vector based on meta data of content, the content vector including one or more aspects of the content;
a storage that stores the content vector and a preference vector that includes one or more aspects of the content that are preferred by a user, whereinthe preference vector is a weighted preference vector in which the one or more aspects of the preference vector each include plural sub-vectors, each sub-vector including a number and a word,the number corresponds to a relative preference of the word with respect to other words in the weighted preference vector, andthe word represents an identifier for a corresponding sub-vector of the preference vector; and
the circuitry further configured to calculate a cosine difference between the one or more aspects of the preference vector with corresponding aspects of the content vector,calculate a degree of similarity between the content vector and the preference vector based on the cosine difference,generate weighted attribute information by weighting each aspect of the content vector based on the degree of similarity,assign a larger weight value to an aspect of the one or more aspects of the content vector having a higher degree of similarity, and select recommended content based on the weighted attribute information.
0 Assignments
0 Petitions
Accused Products
Abstract
A program vector can represent attributes of a program to be generated. EPG data can be received, and metadata necessary for generation of a program vector PP can be extracted from the EPG data. A morphological analysis can be carried out on contents and title included in the metadata to disassemble the contents and the title into words. Items included in the metadata can be subjected to a vector creation process to generate the program vector PP. An effect vector can be extracted on the basis of a genre of a program associated with the metadata. The extracted effect vector can be associated with the generated program vector PP and the processing can be ended. This can be applied to a distribution server for distributing contents.
-
Citations
17 Claims
-
1. An information processing apparatus, comprising:
-
circuitry configured to generate a content vector based on meta data of content, the content vector including one or more aspects of the content; a storage that stores the content vector and a preference vector that includes one or more aspects of the content that are preferred by a user, wherein the preference vector is a weighted preference vector in which the one or more aspects of the preference vector each include plural sub-vectors, each sub-vector including a number and a word, the number corresponds to a relative preference of the word with respect to other words in the weighted preference vector, and the word represents an identifier for a corresponding sub-vector of the preference vector; and the circuitry further configured to calculate a cosine difference between the one or more aspects of the preference vector with corresponding aspects of the content vector, calculate a degree of similarity between the content vector and the preference vector based on the cosine difference, generate weighted attribute information by weighting each aspect of the content vector based on the degree of similarity, assign a larger weight value to an aspect of the one or more aspects of the content vector having a higher degree of similarity, and select recommended content based on the weighted attribute information. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. An information processing method, comprising:
-
generating, by circuitry, a content vector based on meta data of content, the content vector including one or more aspects of the content; storing the content vector and a preference vector that includes one or more aspects of the content that are preferred by a user, wherein the preference vector is a weighted preference vector in which the one or more aspects of the preference vector each include plural sub-vectors, each sub-vector including a number and a word, the number corresponds to a relative preference of the word with respect to other words in the weighted preference vector, and the word represents an identifier for a corresponding sub-vector of the preference vector; calculate, by the circuitry, a cosine difference between the one or more aspects of the preference vector with corresponding aspects of the content vector; calculate, by the circuitry, a degree of similarity between the content vector and the preference vector based on the cosine difference; generating, by the circuitry, weighted attribute information by weighting each aspect of the content vector based on the degree of similarity; assigning, by the circuitry, a larger weight value to an aspect of the one or more aspects of the content vector having a higher degree of similarity; and selecting, by the circuitry, recommended content based on the weighted attribute information. - View Dependent Claims (10, 11, 12, 13, 14, 15)
-
-
16. A non-transitory computer readable medium storing computer executable instructions that, when executed by a computer including a processor, cause the computer to:
-
generate a content vector based on meta data of content, the content vector including one or more aspects of the content; store the content vector and a preference vector that includes one or more aspects of the content that are preferred by a user, wherein the preference vector is a weighted preference vector in which the one or more aspects of the preference vector each include plural sub-vectors, each sub-vector including a number and a word, the number corresponds to a relative preference of the word with respect to other words in the weighted preference vector, and the word represents an identifier for a corresponding, sub-vector of the preference vector; calculate a cosine difference between the one or more aspects of the preference vector with corresponding aspects of the content vector; calculate a degree of similarity between the content vector and the preference vector based on the cosine difference; generate weighted attribute information by weighting each aspect of the content vector based on the degree of similarity; assign a larger weight value to an aspect of the one or more aspects of the content attribute information having a higher degree of similarity; and select recommended content based on the weighted attribute information. - View Dependent Claims (17)
-
Specification