Methodological apparatus for monitoring of the text information content, using semantics of the domain

Monitoring of the subject areas by the exponentially growing text information content is widely used in various fields of social activity. Such areas include industrial development, scientific advances, military and political situations, etc. However, the existing automated systems are characterized by low efficiency of search and text processing respectively, since in most cases the are based on the analysis of lexical structures of various types of their complexity, while the true meaning of the studied documents is not revealed. This leads to significant error in the selection and analysis of the received documents and consequently to lower effectiveness of such systems.

This paper proposes to build the above mertioned systems using the methodological apparatus that allows to make the semantic interpretation of texts by considering additionally such elements of the attributes of text structures as functional role structures, they have the groups of rhetorically related entities (the result of the discourse analysis), as well as the conceptual and substantive universals of object ontology.

The article contains the mathematical tools to calculate the correct completeness of text handling, some original general algorithms from the discourse analysis and multi-agent technologies for processing ontological descriptions, as well as an integrated approach to such treatment.