TRADITIONAL AND INNOVATIVE METHODS OF LINGUISTIC RESEARCH: ANALYSIS OF MODUS CATEGORIES IN MEDIA DISCOURSE

Abstract

The article analyzes linguistic research methods used in contemporary linguistics. It provides a characterization of traditional methods of linguistic analysis, encompassing both general and specialized approaches, and describes the stages of development and key features of modern research techniques based on neural network technologies. Particular attention is given to lexical vectorization techniques, linguistic modeling, and the use of recurrent neural networks. Practical examples of software application (such as a combination and permutation generator and a set operations algorithm) are provided to illustrate the study of modus category functioning in media discourse. A classification of categorical cohesion models is proposed, and instances of semantic interaction and mutual complementarity between various modality categories are identified. The results of the study may serve as a foundation for the development of automated tools for media text analysis.