Method and apparatus for searching video section using natural language

1. A method for searching for a video section, the method comprising:

extracting keywords and a relational operator between keywords from a natural language sentence when the natural language sentence is input, the keywords being for searching a video section;

determining whether or not the extracted keywords are included in predefined context information;

identifying first and second keywords among the extracted keywords, the first keyword being included in predefined context information, and the second keyword being not included in predefined context information;

deriving a first search result by performing a first method on the first keyword, the first method being a method of searching a video section with predefined keyword, object, face, or attribute;

deriving a second search result by performing a second method on the second keyword, the second method being a method of searching a video section corresponding to a sentence inputted by a user; and

deriving a final search result by merging the first and second search results based on the relational operator, and providing the first search result to a user.

2. The method of claim 1 , wherein the first method uses one or more artificial intelligence learning models according to a type of context information.

3. The method of claim 1 , wherein, when the extracted keywords belong to predefined context information, a predefined model which the extracted keywords correspond to is determined.

4. The method of claim 3 , wherein the predefined model is enabled when context information belonging to each model is given.

5. The method of claim 1 , further comprising, when the second method is performed, learning a keyword, for which the second method is performed, as a predefined keyword.

6. The method of claim 5 , wherein, when the learned keyword is input again, the first method is performed instead of the second method.

7. The method of claim 1 , wherein the predefined keyword is a keyword that is stored as a predefined keyword in an internal memory.

8. The method of claim 1 , wherein the predefined keyword is a keyword that is provided by an external device.

9. The method of claim 1 , wherein the providing of the search result provides a search result by prioritizing search results in order of accuracy.

10. An apparatus for searching for a video section, the apparatus comprising:

a memory; and

a processor operatively coupled to the memory;

wherein the processor is configured to:

extract keywords and a relational operator between keywords from a natural language sentence when the natural language sentence is input, the keywords being for searching a video section;

determine whether or not the extracted keywords are included in predefined context information;

identify first and second keywords among the extracted keywords, the first keyword being included in predefined context information, and the second keyword being not included in predefined context information;

derive a first search result by performing a first method on the first keyword, the first method being a method of searching a video section with predefined keyword, object, face or attribute,

derive a second search result by performing a second method on the second keyword, the second method being a method of searching a video section corresponding to a sentence inputted by a user, and

derive a final search result by merging the first and second search results based on the relational operator, and provide the first search result to a user.

11. The apparatus of claim 10 , wherein the processor is further configured to enable a keyword, for which the second method is performed, to be learned as a predefined keyword, when the second method is performed.

12. The apparatus of claim 10 , wherein the processor is further configured to derive a search result by prioritizing search results in order of accuracy.