Methodological approaches to identifying the features of false information within the information security system of the Defense Forces of Ukraine

Abstract

The article substantiates the conceptual and methodological foundations for identifying the characteristics of false information within the information security system of the Defense Forces of Ukraine, emphasizing the growing importance of information security and cognitive resilience in contemporary conflicts. The study systematizes and critically analyzes modern approaches to the detection of disinformation, integrating cognitive, content-semantic, behavioral, technical, and source-based dimensions. It is established that the majority of existing studies address only partial aspects of this phenomenon, lacking a unified methodological framework capable of combining qualitative and quantitative parameters of credibility assessment.
The proposed approach introduces a comprehensive classification of five groups of false information indicators and defines specific quantitative metrics for their evaluation. These include entropy and frequency analysis of textual data, Zipf's law for lexical distribution regularities, TF-IDF weighting for semantic relevance, cosine similarity for intertextual comparison, and the replication rate coefficient to model the dynamics of message dissemination. The integration of these components forms an integral credibility model that enables the transition from descriptive assessments to formalized, measurable evaluation of information veracity.
The developed methodological framework allows for a multi-layered analysis of information content and dissemination patterns, providing a more objective basis for detecting manipulative or disinformative narratives within hybrid influence operations. The results obtained contribute to the advancement of scientific understanding of disinformation mechanisms and create a foundation for the development of automated analytical systems capable of identifying hybrid information threats in real time.
The study's practical significance lies in its potential application for the design of intelligent information monitoring systems and early warning mechanisms against disinformation campaigns. Future research directions include the refinement of weighting coefficients for credibility indicators through machine learning, the extension of the model to multimedia content, and the exploration of cognitive response dynamics as a critical factor in the spread of false information.