DEVELOPMENT OF A MECHANISM FOR DETECTING MANIPULATIVE POTENTIAL OF ADVERTISING MESSAGES IN SOCIAL MEDIA

Abstract

The article is devoted to the development of a mechanism for detecting the manipulative potential of advertising messages in social media. The growing role of SMM management and marketing strategies in social networks has led to the active use of linguistic manipulative tools aimed at shaping consumer attitudes. The article analyzes the main linguistic techniques of manipulation. The aim of the study is to develop a methodology for the automated analysis of advertising texts in social networks to identify linguistic means of manipulative influence and detect signs of manipulativeness in advertising messages using corpus linguistics and modern natural language processing technologies. To achieve this goal, an analysis of linguistic means of manipulative influence was conducted, and a mechanism for identifying signs of manipulativeness in advertising messages in social networks was created using corpus linguistics and information technologies. The statistical method TF-IDF (Term Frequency – Inverse Document Frequency) was applied to identify linguistic units with high manipulative potential. During the analysis, Python tools, particularly the spaCy library, were used to automate the process of detecting manipulative adjectives in a corpus of advertising messages. The scientific novelty of the study lies in the integration of linguistic analysis and natural language processing methods to identify linguistic manipulations in the digital environment. The proposed algorithm can be used to enhance the level of information security for consumers. The obtained results can be useful for consumers seeking to protect themselves from manipulative influence, as well as for SMM specialists aiming to create ethical advertising.