Computational linguistics in psychology: a key to understanding language and human behavior

Abstract

The relevance is determined by the exponential growth of textual data in the digital environment, requiring automated analysis methods to reveal hidden patterns. The aim of the paper is to analyze the possibilities and limitations of computational linguistics methods in personality research. The hypothesis is that automated text analysis can reveal psychological characteristics of authors. Methods and materials: we used the National Corpus of the Russian Language (NCLR) for linguistic frequency analysis, NLP methods (tone analysis, keyword extraction) to determine emotional coloring and identify key themes. Syntactic constructions were analyzed to identify peculiarities of writing style. Results: the analysis of NLP allows to identify the prevailing emotional states and cognitive styles of the authors. Frequent use of first person singular pronouns may indicate egocentricity. Tonal analysis and key word selection allow us to determine the emotional coloring of the text. Analysis of syntactic constructions, such as the preference for passive voice, may indicate avoidance of responsibility. Complex sentences indicate a desire for detail. The novelty of the study is the systematization and adaptation of computational linguistics methods for designing dynamic cognitive systems for diagnosis and non-medical correction. Conclusions: computational linguistics provides powerful tools for personality research, requiring careful and informed application with ethical considerations.