Abstract
The article proposes an apparatus of logic and linguistic modeling for extracting knowledge from electronic text documents. Different types of forms of logic and linguistic models have been singled out for the formalized representation of complex sentences of natural language. The author formulates the rules and carries out the classification of separate forms of logic and linguistic models depending on syntactic constructions of sentences natural language. Such separate forms of logic and linguistic models are templates with certain logical operations, which are used depending on the punctuation marks and connecting words applied in sentences. The specific order of simple predicates depends on the logical connections between simple sentences in the structure of the complex one. This allows to develop an algorithm for further analysis of textual information and its recovery from formal models of knowledge representation. Figs.: 1. Refs.: 12 titles.