The article is devoted to the relevance of the social values of liberal-democratic societies during the war, their structural features, and their role in the post-war reconstruction of Ukraine. The corresponding values include safety, diversity, selectivity, normativity, freedom, equality, order, stability, development, independence, and strength. The empirical part of the study is based on the results of an online survey conducted in July 2022. The sample size is 2,350 respondents (women – 51,3 %, men – 48,7 %, average age – 37,3 years). The structure of the sample repeats the structure of the urban population of Ukraine aged 18 to 55 at the beginning of the war in settlements with 50 thousand inhabitants or more for all regions, except the following groups:              1) population on the territories of Donetsk and Luhansk regions temporarily occupied at the beginning of the war, as well as AR Crimea; 2) population on other territories of Luhansk region; 3) population on the territories of Kherson region. The ranking of societal values (Condorcet method) was established: safety (80,5 %), strength (63,8 %), order (61,3 %), normativity (58,9 %), equality (53,6 %), liberty (48,4 %), stability (44,0 %), independence  (37,4 %), development (35,6 %), diversity (34,1 %), selectivity (32.3 %). Two clusters are identified in the structure of societal values. The first includes order, safety, diversity, selectivity, normativity, liberty, and equality. In general, this value cluster emphasizes the need for internal balance, which is a fundamental guarantee of national resilience during a large-scale war, as it preserves the ability of society to generate additional resources needed by the state. The second value cluster includes strength, development, independence, and stability. This substructure is related to that part of national resilience responsible for responding to external crises and challenges. Forecasts were made regarding the priority relevance of almost all societal values in the period of post-war reconstruction: 1) safety due to the damage caused by the war to the population and territories of Ukraine; 2) strength due to the successes of Ukraine in the defense and diplomatic spheres; 3) order due to the inadmissibility of cleavage narratives in society; 4) normativity due to the request for implementation of the rule of law; 5) equality due to the need for support of broad sections of the population from the state; 6) liberty due to the need of gradual easing of restrictions that arose due to the war; 7) stability due to the constant demand for it in peacetime; 8) development due to its fundamental importance for the national subjectivity of Ukraine;              9) diversity due to modern landmarks of civilizational development;      10) selectivity due to the need for effective management decisions.

У статті розглянуто можливості використання сучасних методів обробки текстів для соціологічного аналізу. Основну увагу приділено трьом завданням, які наразі можна вирішити засобами обчислювального ана­лізу текстів: аналіз змістовної близькості, моделювання тем та сенти­мент-аналіз (аналіз тональностей). В останні роки методи обробки природної мови настільки прогресували, що це дає змогу соціологам автоматично фіксувати семантику текстів, порівнювати її в часі, групувати на підставі схожості. Також це уможливлює масштабування аналізу вели­ких масивів документів, що відкриває нову сторінку в розвитку контент-аналізу, за якої ми наближаємося до відмови від ручного кодування документів, а дослідники зможуть сконцентруватися на аналізі. Ми продемонстрували ці можливості на прикладі аналізу новин із ресурсу «Українська правда» за 2001–2020 рр. Методи, застосовані в статті, дали нам змогу повністю автоматизовано виявити, які семантичні зрушення щодо слів, пов’язаних із діяльністю правоохоронних органів, відбувалися під дією соціальних факторів протягом останніх двадцяти років. Також ми згрупували новини за основними темами повідомлень про поліцію в матеріалах видання й проаналізували, чи змінювалося ставлення до неї протягом його існування.

The latest analysis methods of sentiments borrowed from computational linguistics are relevant in the age of big data, which is difficult to process through traditional content analysis. These methods have made it possible to analyze information over a long period, which allows us to trace the dynamics of the relationship to a particular object over time and large-scale comparative studies of texts. The authors demonstrate the applicability of sentiment analysis based on transformer models to the study of the temporal model of attitudes towards well-known politicians (2001-2021) on the example of text analysis of multilingual online publications. To do this, the authors used the targeted-BERT method for automated directed analysis of sentiments, obtained quality indicators F1-score 0.799 and 0.741 for Ukrainian and Russian models, respectively. The authors tested the dependence of mediatization of politicians on the country's political hierarchy, confirmed hypotheses about the attitude to their power (more significant criticism of the Ukrainian media and gradual loyalty to the Russian media) and foreign politicians (dominance of negative tone in both media with a growing trend for Ukrainian media).

This article suggests a way to concretize the concept of social polarization, which will be most suitable for a) empirical operatio­nalization; b) a complete description of existing social conflicts; c) will be consistent with existing sociological theories. The implementation of this task opens the way to studying this phenomenon through the method of text mining. We see two main problems with the concept of social polarization in sociology: (1) Social polarization is used as a beautiful metaphor to describe contemporary political situations, not as a strong operationalized concept. The concept must create vast opportunities to study social reality, interpret more processes; (2) The mathematic interpretation of social polarization is conducted on somewhat idealized distributions; there is a lack of real empirical data verification. These two problems also create one big problem: mathematical conceptualization of social polarization and empirical studies of social polarization are unrelated. We propose a way to solve this problem through the construction of our social polarization theoretical framework. The way that allowed us to do this was in the concretization of social polarization and its connection with sociological theories of conflict. The article’s key idea is to show that this concept is suitable for operationalization for two reasons: its ability to describe the causes and nature of social conflicts and its measurability. This article also discusses the main modern social polarization theories, their features, advantages, and disadvantages. Since the concept of social polarization is mostly the focus of political science research, the author’s goal was to find opportunities to use this concept in sociology and the ideas that will allow it. There are currently two approaches to studying this issue: the party association approach and the opinion-based group approach. An important task, which was also solved in this article, is the concept’s connection with the sociological concepts of conflict. The path was found using the concept of Lipset and Rokkan. This concept’s key advantage is the combination of social inequality, conflict, attitudes, and social distance. Typically, these concepts are used separately to explain social cleavages. The concept of polarization, in this case, allows them to be integrated into a single whole