Server and method for providing content based on context information
AssigneeKorea University Research and Business Foundation
InventorsSang Keun Lee, Woo Jong Ryu
Life of this patent
Estimating the term…
Abstract
The present disclosure includes a memory in which a content providing program is stored, and a processor that executes the program stored in the memory. Herein, upon execution of the program, the processor extracts a subject and a verb property including multiple verbs corresponding to the subject from a target content, infers a verb-form subject of the target content on the basis of the subject and the verb property, and extracts one or more contents matched with the target content from among multiple candidate contents on the basis of the verb-form subject of the target content. Further, the multiple verbs corresponding to the subject are extracted from an associative verb group.
Claims12
A server for providing a digital content based on context information, comprising:
- a memory in which a digital content providing program is stored;
- and
- a processor that executes the program stored in the memory,
- wherein upon execution of the program, the processor extracts one or more verbs from a digital content for each subject included in a subject classification tree, matches the each subject with the one or more verbs extracted from each subject to generate an associative verb group, extracts a subject from a target digital content, extracts one or more verbs corresponding to the extracted subject from the associative verb group, extracts a verb property representing weight values in vector form which are calculated based on similarity between the target digital content and the extracted one or more verbs from the associative verb group,
- infers a verb-form subject of the target digital content on the basis of the extracted subject and the verb property, and
- extracts one or more digital contents matched with the target digital content from among multiple candidate digital contents on the basis of the verb-form subject of the target digital content, and
- wherein the verb property is defined by the following Equation 1:
- {right arrow over (vfp,c)}=(w1,w2, . . . wm) <Equation 1>
- where p represents the target digital content, c represents the extracted subject from the target digital content, {right arrow over (vfp,c)} represents the verb property, w1 represents the weigh value for the first verb included in the verb property, w2 represents the weigh value for the second verb included in the verb property, and wm represents the weigh value for an mth verb included in the verb property;
- and
- wherein the verb-form subject is provided in the form of a pair of the extracted subject and the verb property corresponding to the extracted subject.
A method for providing a digital content based on context information by a server for providing a digital content, comprising:
- extracting one or more verbs from a digital content for each subject included in a subject classification tree;
- generating an associative group by matching the each subject with the one or more verbs extracted from each subject;
- extracting a subject from a target digital content;
- extracting a verb property represents weight values in vector form which are calculated based on similarity between the target digital content and the extracted one or more verbs from the associative verb group;
- inferring a verb-form subject of the target digital content on the basis of the extracted subject and the verb property;
- and
- extracting one or more digital contents matched with the target digital content from among multiple candidate digital contents on the basis of the verb-form subject of the target digital content,
- wherein the verb property is defined by the following Equation 1:
- {right arrow over (vfp,c)}=(w1,w2, . . . wm) <Equation 1>
- where p represents the target digital content, c represents the extracted subject from the target digital content, {right arrow over (vfp,c)} represents the verb property, w1 represents the weigh value for the first verb included in the verb property, w2 represents the weigh value for the second verb included in the verb property, and wm represents the weigh value for an mth verb included in the verb property;
- and
- wherein the verb-form subject is provided in the form of a pair of the extracted subject and the verb property corresponding to the extracted subject.
Dependencies are inferred from the claim text.