Bisikalo O. Information technology of image analysis of text and synthesis of structure-functional models of system of natural-language contents processing

Українська версія

Thesis for the degree of Doctor of Science (DSc)

State registration number

0513U000027

Applicant for

Specialization

  • 05.13.06 - Інформаційні технології

21-12-2012

Specialized Academic Board

Д 05.052.01

Vinnytsia national technical university

Essay

The object of research is the process of word processing, imagery analysis and synthesis of structural and functional models of systems based on natural language information; the purpose of the study is to improve the quality of the solution of semantic-dependent tasks of information processing by creating a methodology and information technology of figurative text analysis and synthesis of structural and functional models of processing of natural language content (PNLC), which focuses on general knowledge of figurative sense, capable of self-improvement and make your own set of functions provides the solution of a complex of semantic-dependent tasks; methods of research are based on use of the system analysis, algebra of logic, fuzzy mathematics, theories of sets, formal systems, information and algorithms, graph theory; theoretical results - to solve the problem of creating a new class of information technology and systems of PNLC the basic theory of figurative analysis of the textual information and the synthesis of structural and functional models of PNLC are developed, including the first time: an approach to the figurative analysis of textual information , based on the concepts of the formal introduction infological systems, ontogenetic principle and linguistic image, which allows numerically evaluate the common figurative sense of natural language constructions is offered, the methodology of synthesis of structural and functional models infological system where human intellectual activity interpreted as an interconnected data processing function of 9 blocks-models of the brain through 19 figurative the nature of information flows, which provides a mechanism of formalizations of PNLC is developed, complex of models and methods of PNLC, allowing to unify data handling processes and knowledge systems to solve a basic and applied semantic-dependent tasks is offered; practical results - reducing time spent experts involved to resolve semantic-dependent tasks, increasing the relevance of full-text search in terms of accuracy and completeness, improved operational performance knowledge base and other components PNLC systems, increase of success of students who being trained independently. The degree of implementation - the results obtained in the thesis, introduced at enterprises "InnoVinn" (Vinnytsia), "AETHER. Inc "( USA)," Smart.exe Ltd "(Israel), 2 state research institutions of Ukraine, public organization, department of education and science of the Vinnytsia Regional State Administration, as well as in the educational process of the specialized educational school and 6 universities in Vinnytsia, Kharkov, Sumy, Kamenetz-Podolsk and Poltava. Sphere (industry) use - systems and semantic-dependent tasks of PNLC.

Files

Similar theses