EXTRACTION OF INFORMATION FROM SANITARY BLOCKS OF DOCUMENTS USING MICROMODELS ON BASIS OF ONTOLOGY Russian patent published in 2018 - IPC G06F17/27 

FIELD: data processing.

SUBSTANCE: invention generally refers to the processing of texts in natural language, and in particular to the extraction of information from the semantic blocks of documents using micromodels based on ontology. In the method of extracting information from documents containing text in natural language, a semantic block belonging to a given category is identified in the text. Perform a lexical analysis of the set of words of the semantic block with the goal of constructing a set of lexical structures containing information about the lexical meanings of words and the corresponding semantic classes representing the semantic block. Identify the micromodel for retrieving information related to a given category, the micromodel including a plurality of product rules associated with the ontology. Apply the production rules of the micromodel in order to extract information objects related to the corresponding semantic class, corresponding to the concept of ontology.

EFFECT: technical result is an increase in the speed and quality of information extraction by using ontology micromodels for individual parts of the document.

22 cl, 13 dwg