FIELD: physics.
SUBSTANCE: to evaluate the text classifier parameters based on semantic characteristics, the semantic-syntactic text analysis in natural language from the body of texts in natural language is performed using the processing device to create a semantic structure representing a set of semantic classes. The text characteristic in natural language is identified, extracted based on a set of values from a set of the characteristic extraction parameters. The body of texts in natural language is separated into a training data sample including the first set of texts in natural language, and a test sample including the second set of texts in natural language. A set of parameter values is defined for extracting characteristics, taking into account the category of the training sample. The obtained set of parameter values is evaluated for extracting characteristics using the test sample.
EFFECT: improving the accuracy of classification results.
20 cl, 15 dwg
Title | Year | Author | Number |
---|---|---|---|
CLASSIFICATION OF TEXTS ON NATURAL LANGUAGE BASED ON SEMANTIC SIGNS | 2016 |
|
RU2628436C1 |
TRAINING CLASSIFIERS USED TO EXTRACT INFORMATION FROM NATURAL LANGUAGE TEXTS | 2018 |
|
RU2691855C1 |
CLASSIFIER TRAINING USED FOR EXTRACTING INFORMATION FROM TEXTS IN NATURAL LANGUAGE | 2018 |
|
RU2681356C1 |
RETRIEVAL OF INFORMATION OBJECTS USING A COMBINATION OF CLASSIFIERS ANALYZING LOCAL AND NON-LOCAL SIGNS | 2018 |
|
RU2686000C1 |
EXTRACTING INFORMATION OBJECTS WITH THE HELP OF A CLASSIFIER COMBINATION | 2017 |
|
RU2679988C1 |
USING VERIFIED BY USER DATA FOR TRAINING MODELS OF CONFIDENCE | 2016 |
|
RU2646380C1 |
MULTI STAGE RECOGNITION OF THE REPRESENT ESSENTIALS IN TEXTS ON THE NATURAL LANGUAGE ON THE BASIS OF MORPHOLOGICAL AND SEMANTIC SIGNS | 2016 |
|
RU2619193C1 |
NAMED ENTITIES FROM THE TEXT AUTOMATIC EXTRACTION | 2014 |
|
RU2665239C2 |
METHOD OF EXTRACTING FACTS FROM TEXTS ON NATURAL LANGUAGE | 2016 |
|
RU2637992C1 |
SENTIMENT ANALYSIS AT THE LEVEL OF ASPECTS USING METHODS OF MACHINE LEARNING | 2016 |
|
RU2657173C2 |
Authors
Dates
2017-08-16—Published
2016-04-12—Filed