Abstract: The article is devoted to models theoretical approaches for detecting and processing contradictions in document texts. Contradictions arising in documents are divided into three types: logical, ontological and substantive. Logical contradictions arise when documents contain statements about regulations that are contradicted by descriptions of specific domain keys. Ontological contradictions arise when the information presented in a document or set of documents is in conflict with the ontology of a given subject area. The developed approach uses ontological homomorphisms and submodels of partial models. When identifying substantive contradictions, it is necessary to extract formal descriptions of situations from documents. The use of the mechanism of partial models is necessary for the formalization of situations, while the formalization of relations between situations is based on the use of ontological homomorphisms of partial models. The proposed methods are based on the construction of chains of the above-mentioned homomorphisms. When processing substantive contradictions, it is necessary to model dynamics using special constants-situations.

Cite: Palchunov D.E. , Chernyavtseva S.I.
Methods for Identifying and Resolving Contradictions in Document Texts
In compilation Proceedings 2024 IEEE International Multi-Conference on Engineering, Computer and Information Sciences (SIBIRCON), Novosibirsk, Russian Federation, 2024. – IEEE., 2024. – C.327-332. – ISBN 979-8-3315-3202-4. DOI: 10.1109/sibircon63777.2024.10758488 Scopus OpenAlex

Dates:

Published print:	Oct 9, 2024
Published online:	Oct 9, 2024

Identifiers:

Scopus:	2-s2.0-85212058640
OpenAlex:	W4404740703

Citing: Пока нет цитирований

Altmetrics: