Study of Gender Stereotypes on the Material of the Russian National Corpus

Social research based on big text data is an interdisciplinary field that combines methods and approaches from the social sciences, computational linguistics, and digital humanities. Traditional data collection methods, such as surveys and interviews, are limited in resources. In contrast, big text data cover a wide range of sources, allowing researchers to analyze millions of texts written by different groups of people around the world. They include information from social media, digital libraries, news sites, and other online sources. This paper also belongs to the field of sociological research based on big data and analyzes the hot topic of gender stereotypes. Gender stereotypes are deep-rooted and widespread beliefs about roles, behavioral norms, preferences, and personality characteristics that are attributed to individuals based on their gender. Such prejudices often negatively affect specific areas of an individual's life and shape norms of behavior, social roles, career paths, and expectations about physical appearance. For example, gender stereotypes limit opportunities for personal and professional growth by assigning certain roles and responsibilities to men and women. Undoubtedly, parents and close environment have a significant influence on the formation of attitudes and views on the life of an individual. However, this process is a complex set of multiple factors, the qualitative and quantitative measurement of which is a significant challenge. At the same time, the information published by the media is available for analysis. In the present study, it was decided to use distributive semantics as a tool for analyzing big data represented by newspaper media publications. The methods of interpretation of vector representation of words allow to identify implicit attitudes and stereotypes. The author of this paper decided to focus on gender stereotypes regarding professional roles. The paper examines the positions of vectors of words denoting occupations. The decision to focus on this type of stereotypes was made on the basis of the fact that words denoting professions have an interesting specificity, there are forms of words used only in relation to women - feminitives. The origin of feminitives may be due to sociocultural and linguistic trends prevailing in society. In some contexts, the formation of feminitives reflects the attitude to gender roles and stereotypes characteristic of a given society at a certain historical moment. Such specificity of the analyzed data requires special attention, because according to the hypothesis put forward, feminitives should show a connection with the female gender, at the same time, the interpretation of the position of masculine words in the vector space can be ambiguous. In recent years, there has been an increasing tendency to use feminitives to refer to women in the publications of the Russian mass media. Thus, the corpus of newspaper media seems to be an interesting object for research.