Digital Communication of the Russian Political Parties: Structure and Content Features in the Eve of All-Russian Referendum – 2021 Elena Brodovskayaa, Anna Dombrovskayaa and Dmitry Karzubovb a Financial University under the Government of the Russian Federation, 49/2 Leningradsky Ave., Moscow, 125167, Russia b Moscow Government, 13 Tverskaya St., Moscow, 125032, Russia Abstract The paper concerns to the analysis of political parties’ digital communications structure in the pre-electoral cycle of 2021. All-Russian Referendum – 2021 will discover public support potential for the current Russian social and political courses. There is a huge possibly for the State Parliament to be changed, the political balance in the State Duma seems to be restructured. The study focused on the automated analysis of the digital markers of the parties’ communication and their regulation strategies in driving online network communities. The methods of data collection and analysis are the developing of social graphs (the structure and density of user connections within social media communities) via proprietary software and cognitive mapping of political party digital content with the creation of a database and its analysis via the SPSS Statistics 24.0. The cases of empirical analysis were the political parties of the Russian Federation - participants of the Referendum – 2020. The key outcome of the study – segmentation of all online political party groups (analysis cases) into three clusters: groups regulated by online macroleaders (more than 100 leaders with more than followers in a digital group); communities managed by online microleaders (less than 100 leaders with more than 100 followers in a digital group); groups with "horizontal connections" that do not have influential influencers. For each type of community, the features of the social media communication content are defined. In conclusion, recommendations on the organization of social and media support for the activities of political parties were formulated. Keywords 1 Social media, digital communications, political parties, social graphs, online community management strategies, cognitive map of party communication 1. Introduction Social Media Management is a key challenge in the development of modern political parties seeking to scale electoral support through digital communications and trying to raise target audiences on social media. The process of Russian political parties’ adaptation to the digital age is extremely uneven. The greatest efforts in this area made, as a rule, by parliamentary parties with a high level of electoral potential, as well as by party actors who initially, at the time of their creation, relied on "digital citizens". A special place in this process belongs to opposition party and pro-party formations, which use the capabilities of their digital infrastructures mainly in order to mobilize protest sentiments of social media users. At the same time, a common problem for political parties with different statuses and spectral affiliations is the problem of effective digital communications with external target groups, since in most cases, digital infrastructures are aimed primarily at solving the problems of retaining the nuclear electorate and communication with activists and supporters. IMS 2021 - International Conference "Internet and Modern Society", June 24-26, 2021, St. Petersburg, Russia EMAIL: brodovskaya@inbox.ru (A. 1); an-doc@yandex.ru (A. 2); karzubovdn@gmail.com (A. 3) ORCID: 0000-0001-5549-8107 (A. 1); 0000-0002-2015-2349 (A. 2) © 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org) 326 PART 2: E-Governance The second common problem for various political parties in the Russian Federation is the limited use of effective practices for converting users' digital activity into voices. The solution to these problems is in direct proportion to which strategy for managing digital communities chosen and consistently implemented by party actors. 2. Theoretical Review Research areas devoted to Internet communications of political parties are as follows. The difficulties of adapting political parties to the digital needs and expectations of the electorate reflected in the largest group of works: D. Janssen [1], U. Klinger [2], G. Mascheroni [3], etc. The use of digital communications by political parties to resolve electoral tasks is the focus of R. Karlsen’s studies [4]. The concept of a hypermedia campaign introduced in D. Lilleker’s and coauthors' publications [5]. B. Kalsnes identifies three disadvantages when communicating with voters on the Internet: reputation risk, negative media attention and limited resources [6]. N. Browning, K.D. Sweater explore how the media agenda, party frames, candidates' traits and communication between a political organization and the public affects for the party reputation [7]. One of the leading directions of modern scientific research in the field of studying social media influence on the development of party systems and political parties is that party actors are better adapted to the conditions of post-materialist political culture due to digital communication technologies. Thus, A. Chadwick [8] hypothesizes that parties renewed under the influence of the digital environment, since citizens involved in Internet communication transform the format of the traditional party organization, rejecting the norms of hierarchical discipline and habitual party loyalty. The idea that digitalization contributes to the formation of a new party model is supported by P. Gerbaudo, thanks to whom the concept of “platformization” was further developed. The essence of this concept is that in the digital age, political parties repeat the organizational principles of digital social networks to scale up electoral support [9]. The most relevant foreign works relevant to the research topic also indicate the need to improve the efficiency of information flow management in social media. In particular, N. Browning focuses on how the information agenda, frames and brands in social media affect the electoral potential of political parties. Researchers emphasize the need to manage information flow and party Internet content to effective use digital communications with targeted groups. In their opinion, dysfunctions in this area course risks of an overflow of electoral votes and a decrease in the stability of the electoral base of political parties. 3. Research Methodology The research methodology based on the following approaches: new institutionalism in its historical and sociological versions [10], [11], as well as network approach [12]. The new institutionalism not only allows one to describe the nature of stability / instability of the positions of actors in the party system (the historical version of the approach), but also to take into account, when analyzing the electoral potential of political parties, the search for a balance between formal and informal components in the process of brand promotion in social media (sociological version of the approach) [13], [14], [15]. The application of the network approach in this study is due to the need to solve the problem of identifying the specifics of managing the network structures of political parties in social media [16]. Based on graph theory, software was used as a network analysis tool that allows identify such parameters of networks as size, density, uniformity, strength of connectivity, stability, etc. The presented research is based on social - media predictive analysis (hereinafter – SMPA). SMPA [17] is a research approach based on intellectual search and analysis of digital traces in social media for modeling and forecasting socio-political processes. In the Russian Federation, this direction is developing mainly by the efforts of two research groups: the team of E.V. Brodovskaya [18] (Financial University under the Government of the Russian Federation) and the team of A.S. Akhremenko [19] (National Research University Higher School of Economics). IMS-2021. International Conference “Internet and Modern Society” 327 SMPA is applicable for monitoring, diagnostics, scenario and forecasting of processes:  legitimation / delegitimization of a decision / actor / institution;  mobilization / demobilization of actions of collective actors;  escalation / de-escalation of the conflict interaction between the subjects of the conflict;  transition of online activity to offline; - the choice of strategies for the behavior of target groups;  changes in social well-being, moods of large audiences, etc. Research methods. Strategies for managing social media groups of political parties were analyzed in the process of constructing and interpreting social graphs of online network communities. The graphs reflecting the density and nature of connections within digital groups were created using the author's software "Social Graph", an attachment to the blog hosting "VKontakte". The sample set of the study consisted of 98 regional social media communities of Russian political parties, the following parties became cases - participants of the Referendum – 2020: “United Russia” (hereinafter: UR), “Liberal-democratic party of Russia” (hereinafter: LDPR), “Communist Party of the Russian Federation” (hereinafter: KPRF), “Spravedlivaya Russia” (Fair Russia), (hereinafter: SR), “Communists of Russia” (hereinafter: CR), “Party of Pensioners” (hereinafter: PP), “Rodina” (“Motherland”), (hereinafter: “Rodina”), “Party Rosta” (“Party of Growth”), (hereinafter: PR), “Patriots of Russia” (hereinafter: Patriots), “Yabloko” (Apple), “Novie Lyudy” (“New People”), “Za Pravdu” (“For Truth”), “KPSS” (“Communist Party of the Social Fair”), “Zelenie” (”The Greens), “Parnas” (“Party of people liberty”), “Party Pryamoi Democratii” (The Direct Democracy Party), and the “Russia bez Corruptzii” (“Russia Party without Corruption”). The selection was carried out according to the principle of continuous research: all digital groups of the VK network belonging to the regional branches of the listed political parties became analytical cases. The content of party’s communication in social media with the target audience was studied using the method of cognitive mapping of social media documents of Russian political parties. The calculation of the sample population was based on a systematic sampling method. The same parties that participated in the Referendum–2020 were selected as research cases (see above). For each of the listed parties, 2 of the most numerous social media communities of the official and unofficial type in various electronic social networks were selected: VK, Odnoklassniki, Facebook, Twitter, Instagram. For each community, the step of selecting messages with depth was calculated in the period from 08/01/2020 to 10/10/2020. The total number of analyzed documents was 1600. The logic behind the analyzed period was the possibility to study the specifics of political communication of parties with supporters in the pre-electoral, electoral and post-electoral intervals. 4. Research Results Strategies for managing social media groups of political parties were analyzed via social graphs of online network communities. The graphs reflecting the density and nature of connections within digital groups were created using the author's software "Social Graph", an attachment to the blog hosting "VKontakte". The sample set of the study consisted of 98 regional social-media communities of political parties: UR (the average number of online group subscribers is 52,558), LDPR (the average number of online group subscribers is 14,195), KPRF (the average number of online group subscribers – 26,864), SR (average number of online group subscribers – 3,787), CR (average number of online group subscribers – 3,041), PP (average number of online group subscribers - 616), “Rodina” (average number of online group subscribers - 2412), PR (average number of online group subscribers – 3227), “Patriots” (average number of online group subscribers – 673), “Yabloko” (average number of online group subscribers – 2462), “Novie Lyudy” ( average number of online group subscribers – 2483), “Za Pravdu” (average number of online group subscribers - 5132, KPSS (average number of online group subscribers – 561), “Zelenie” (average number of online group subscribers – 1007), “Parnas” (average number of online group subscribers – 5295), “Party Pryamoi democratii” (average number of online group subscribers – 914). The selection was carried out according to the principle of continuous research: all digital groups of the VK network belonging to the regional branches of the listed political parties became analytical cases. 328 PART 2: E-Governance The strategy of regulating an online group using macrosocial groups within the community is the most rarely implemented in the process of managing political parties’ social media communities – Fig. 1. Only 13.1% of all analyzed digital communities showed signs of having large, influential online influencers numbering at least 100 and with an audience of more than 100 people. These subgroups within the digital community provide a high level of density of connections of the entire online group and a significant level of mobilization potential. The political party leading for this strategy is “UR”: about half of all social media communities (46.2%) identified as groups with a large number of influential influencers belong to this political party. The second position is taken by “KPRF”: 30.8% of its communities belong to the considered segment of online groups. Figure 1: A typical graph of digital groups with a strategy: “Information impact using macrosocial groups within the community” A more common strategy for managing social media political communities is the formation of many relatively small subgroups (from 60 to 100 people), each of which is led by one microleader (36.4% of all analyzed social media groups) – see Fig. 2. Such a community has an average density of connections and significant, but not extreme, potential for mobilizing political / civic activity. More than half of the social media groups (55.6%) in this cluster belong to two parties – “LDPR” and “KPRF” (27.8% each). Figure 2: Typical graph of digital groups with a strategy: “Information impact using microsocial groups within the community” Most often, political parties implement in social media a strategy of distancing themselves from purposeful and systematic work to strengthen ties between users of their communities see Fig. 3. More IMS-2021. International Conference “Internet and Modern Society” 329 than half of the analyzed online groups (50.5%) ignore the technology of increasing the cohesion of the social media community and do not have influential microgroups with topic-starters that purposefully form the value and behavioral attitudes of community members. The political party “Yabloko” has the largest number of such communities (20.0%) of the entire analyzed segment of communities, and this strategy is also common among the online groups of “UR” (18%). A third of all communities in this cluster (30.0%), which are groups of non-parliamentary parties (“KPRF”, “Za Pravdu”, “PR”, “Rodina”), implement the strategy of neglecting systematic work with supporters in social media. Figure 3: Typical graph of digital groups with a strategy: “Lack of regulation of the functioning of the community” The structural characteristics of the relationship between online network leaders and representatives of the user audience of digital communities must be supplemented with a meaningful focus of party communication with the target audience. This will make it possible to fill with contextual meaning the techniques that political influencers use to shape the electoral attitudes and citizenship of Russians. The social media content of party digital groups was studied using the cognitive mapping method. The cognitive map of online network party communication has shown that in the genre palette of messages from parties whose digital communities are most often regulated by means of informational influence using macrosocial or microsocial groups within the community: “UR”, “KPRF” and “LDPR” and “Novie Lyudy” are distinguished by the most non-standard content that goes beyond the typical genre repertoire. From a fifth to a third of the posts of their communities are materials of a wide variety of types: from the activities of the regional branches of the party, their successes in the field to legislative initiatives put forward by the party and the experience of tackling specific social issues in local territories. Biographical, diffuse and social technology-oriented posts should be recognized as literally isolated ones. At the same time, during the election campaign and in the post-election period, even the most successful parties in terms of social media representation ignore the genre of “life stories”, congratulatory content and the presentation of new legislative initiatives among the target audience are also rare. According to the results of the analysis, the prevailing context of posts in digital communities regulated by leaders of macro- and microsocial groups is the problem of social welfare. Most of these party communities’ stake on discussing issues of social support for the population, obviously counting on the approval of broad strata of Russian citizens with low incomes and in need of a strong centralized system of social assistance to the population. Most often, among the groups – research cases, the context of expanding social care is articulated in the “UR” social media posts (45.1%) and “LDPR” (37.5%). Another discourse dominant of communities ruled by macro and micro social influencers is the foreign 330 PART 2: E-Governance policy context, problems of Russia's geopolitical status, and its role in the development of the world political process. In this aspect, “KPRF” party dominates with 10.9% of posts on this topic. At the same time, this dimension of the electoral message to voters should be recognized as rather unpopular, since in the context of a general decrease in incomes and the quality of life of Russians during the pandemic, the demand for overcoming the inner social problems of Russian society has clearly increased, which is fully reflected in the thematic repertoire of social media communication parties during the pre- election period. An important component of the political parties’ digital communications cognitive map is the method of influencing the target audience. In the analyzed online groups rational argumentation prevails: “UR” (67.3%) and a combination of methods of emotional and rational influence: “Novie Lyudy” (43.8%), “KPRF” (48.7%) and “LDPR” (44.8%). Most often, party digital groups try to combine affective and rational influence to achieve the effect of the broadest targeting of their messages in social media. The achievement of the desired effect is confirmed by the data of correlation analysis (Pearson Correlation coefficients: -0.165 * – for likes, – 0.123 ** – for comments, - 0.216 ** – for reposts). These coefficients mean that parties that use predominantly rational reasoning attract the least user resonance, and parties that rely on emotional content generate the most intense user activity. Some social media groups, regulated by online "micro- and macro" leaders, quite often use a persuasion technique called "dehumanization", the nature of which is to dehumanize the opponent, denying the possibility of perceiving this or that political figure – the object of dehumanization from the standpoint of universal human morality. In a number of cases, the “Novie Lyudy” (17.5%) and “KPRF” (30.8%) parties are inclined to this method of fighting ideological opponents. A special place in the opposition parties” list is occupied by methods of emotional and rational stereotyping, a similar principle of which is the formation of social prejudice against the opponent using emotionally resonant characteristics (biography, personal qualities) or false argumentation in relation to the statements / actions of a political opponent. Emotional stereotyping is used by the “KPRF” (30.8%) and “Novie Lyudy” (30.1%). Rational stereotyping is also most often used by the “Novie Lyudy” (22.5%) and “KPRF” (15.4%). The cognitive map of party digital communications also includes an analysis of the trigger, the event context of interaction with voters. For party communities ruled by "micro and macro" leaders, there is a special trigger around which social media connection with supporters and potential audience is built. For example, the trigger for the arrest of Sergey Furgal (ex-Governor of Khabarovsk Region), one of the representatives of this political party. The human rights stream, formed by the “LDPR” party around the former Governor of the Khabarovsk Region, obviously has the goal of opposing the negativization of the image of a politician affiliated with the party. Party communities that do not have influential online leaders and function on the principle of forming horizontal ties have a fundamentally different content of online communication, most often these are groups of the following parties: "SR", "Za Pravdu", "Yabloko", "Zelenie", "Rodina", "Patriots", "PR", "CR", "Parnas". The greatest focus on political analytics for their online supporters is “Parnas” (60.0%), “Rodina” (53.6%), “Zelenie” (52.9%). Slightly less than half of all-party digital content falls on the analysis of political processes in the communities of “SR” (49.4%) and “PR” (49.3%). Part of the communities in the cluster under consideration focuses on the coverage of the events held. The four leaders in this indicator are the “Patriots” (56.7%), “Zelenie” (41.2%), “Za Pravdu” (40.1%) and “Parnas” (40.0%). Based on a quantitative analysis of the party social media content, most often in online communities for their followers, parties post is about meetings of activists and party representatives. The exceptions are “Zelenie” and “Yabloko”, which focus on the conducted actions aimed at constructing a certain social / socio-political / environmental problem. Thus, a huge part of the systemic Actors handling opposition, whose communities are most often focused on the formation of horizontal ties, is aimed at the representation in new media of political analytic content, focused on explaining the supporters of their position on certain socio-political issues. As before, reporting materials remain one of the dominants of the party content in this segment, with only the most opposition-oriented parties focusing not on party reporting conferences, but on held rallies, rallies, and flash mobs. IMS-2021. International Conference “Internet and Modern Society” 331 The leader in terms of the share of messages on the development of entrepreneurship, industry, and the economy as a whole is the “PR” party (52.2%), this context is almost twice as rare in the "Party pryamoi democratii" – 29.5% and 23.3% in "Patriots” party, other parties either do not ignore this type of message to voters at all, or represent it extremely rarely. Thus, it is quite obvious that political parties came to the next electoral cycle with confidence in the demand for ideas of the "left ideology" in connection with the growing impoverishment of the population and a shrinking layer of voters for whom the development of entrepreneurship is an important argument to join the party's supporters. Personal rights defense is a dominant topic for opposition party’s social media content: “Zelenie” (70.6%), “Parnas” (45.0%) and “Yabloko” (25.2%). It testifies to their exploitation of triggers for mobilizing protest moods in connection with the so-called “resonant cases of human rights violation” by the current Russian government. “Parnas” (35.0%) and “Rodina” (31.4%) parties devote digital content to the foreign policy context. Noteworthy is the fact that a big share of party communities’ of "horizontal ties" social media content does not contain any problem at all, these posts do not indicate any contradiction in social life, which may be perceived by the audience of digital messages as emotionally close. Most often, without mentioning the problems to which the attention of a political party will be directed, the posts of the party "Za Pravdu" (65.2%), "PP" (44.7%) and "Yabloko" (39.0%) are published. A significant proportion of party communities with "horizontal ties" prefer such a strategy of value influence on the target audience as "primitivization." Its essence is in the oversimplification of the perception of this or that event / decision / personality of the politician. This is an effective technique for undermining confidence in the political force in power. On a statistically significant scale, such parties as the “PP” (50.1%), “Yabloko” (26.1%), “Rodina” (22.3%). According to the preferred methods of influence, the communities of the analyzed cluster are characterized by the dominance of affective influence on the social media audience: “Patriots” (60.0% of emotional posts), “Rodina” (59.0%), “PP” (57.4%), “CR” (50.0%). These communities of "party outsiders" use the most effective in terms of obtaining resonance – the technique of affective influence. He focuses on emotionally strong messages about social injustice, social inequality and the enormous stratification of Russian society. Emotional visualization of the plight of Russians socially unprotected is a basic technique for attracting supporters, potential voters in these digital groups. The most popular triggers in the "horizontal link" communities for the groups of the “Yabloko” party is the case of Alexey Navalny, for “Parnas” – the Belarusian crisis. The exploitation of the situation with Alexey Navalny and the distortion of its essence were quite intensively used by the communities of the Yabloko party during the Referendum-2020 election campaign and became more frequent in the beginning phase of the State Duma Election – 2021 campaign after A. Navalny returned to Russia and during the organization of the All-Russian protests against his arrest (40.7%). The context of the use of the Belarusian trigger by the online groups of “Parnas” is the idea of the Belarus scenario replication in Russia. 5. Conclusion Summarizing the research results note that the strategy of maintaining high cohesion of the user audience can be implemented only by two parliamentary political parties – United Russia and the Communist Party of the Russian Federation, while the absolute number of regional digital communities of these parties with high mobilization potential and significant cohesion is rather small. The Russian Liberal Democratic Party social media communities and part of the digital groups of United Russia and the Communist Party of the Russian Federation are aimed at retaining the attention of their target groups in the process of informational touches of topic-starters within the framework of small subgroups. From a meaningful point of view, the digital communication of these online groups is focused on social issues, foreign policy agenda and involves the use of rational argumentation. The social media communities of such a parliamentary party as “Spravedlivaya Rossia” (“Fair Russia”) and the majority of non-parliamentary political parties (“Yabloko” (“Apple”), “Za Pravdu” (“For truth”), “PR” (“Party of Growth”), “Rodina” (Motherland”) are characterized by the lowest density of social connections. These online groups are characterized by "horizontal connections", a lack of influential influencers in the community. These groups focus on opposition discourse, exploit triggers 332 PART 2: E-Governance of high-profile human rights cases, and focus on the emotional manner in representation of social media content. In general, the rather low level of political parties’ social media management in pre-electoral period is revealed. It increases uncertainty in predicting the results of voting for candidates for deputies of the State Duma in 2021. At the same time, the traditionally strong positions of “United Russia”, the “Communist Party of the Russian Federation” and, in part, “Russian Liberal Democratic Party” are reinforced by the systematic work of these parties with their user audience. The rest of the parties need to activate the social media management strategy from the standpoint of strengthening the density of connections in the structure of digital communities and intensifying targeted, targeted and regular social media contacts. 6. Acknowledgements This work has been supported by the Russian Foundation for Basic Research (RFBR) and Expert Institute for Social Research (EISR) grant (project No. 21-011-33015). 7. References [1] D. Janssen, R. Kies, Online Forums and Deliberative Democracy, Acta Politica. 40(3), 2005, pp. 317-335. doi:10.1057/palgrave.ap.5500115. [2] U. Klinger, Mastering the Art of Social Media, Information, Communication and Society, 16(50), 2013, pp. 717-736. doi:10.1080/1369118X.2013.782329 [3] G. Mascheroni, A. Mattoni, Electoral Campaigning 2.0 – The Case of Italian Regional Elections, Journal of Information Technology & Politics, 10(2), 2013, pp. 223–240. doi: 10.1080/19331681.2012.758073 [4] R. Karlsen, Velgernes valgkamp. Det politiske landskapet: En studie avstortingsvalget 2009. The Political Landscape. A Study of the National Election 2009. Oslo: Cappelen Damm Akademisk 2011. pp. 41-64. [5] D. G. Lilleker, J. Tenscher, V. Štětka, Towards, Hypermedia Campaigning? Perceptions of New Media’s Importance for Campaigning by Party Strategists in Comparative Perspective, Information, Communication and Society, 18(7), 2015, pp. 747-765. doi: 10.1080/19331681.2017.1397239 [6] B. Kalsnes, The Social Media Paradox Explained: Comparing Political Parties’ Facebook Strategy Versus Practice, Social Media + Society, 2(2), 2016, pp 1-11. doi: 10.1177/2056305116644616 [7] N. Browning, K.D. Sweetser, How media diet, partisan frames, candidate traits, and political organization-public relationship communication drive party reputation, Public Relations Review, 2 (46), 2020, pp. 1-17. doi: 10.1016/j.pubrev.2020.101884 [8] A. Chadwick, J. Stromer-Galley, Digital Media, Power, and Democracy in Parties and Election Campaigns, International Journal of Press/Politics, 21, 2016, pp. 1-11. doi: 10.1177/1940161216646731 [9] P. Gerbaudo, The Digital Party: Political Organization and Online Democracy. Pluto Press 2019 [10] P. Hall, R. Taylor, Political Science and the Three New Institutionalisms, Political Studies, 44, 1996, pp. 36-57. URL: https://www.mpifg.de/pu/mpifg_dp/dp96-6.pdf [11] M. Marcussen, Ideas and Elites: The Social Construction of Economic and Monetary Union, Aalborg University Press 2000. [12] G.V. Gradoselskaya, Network measurements in sociology. Moscow 2004 (In Rus.) [13] P.J. DiMaggio, W.W. Powell, Introduction. The new institutionalism in organizational analysis / W.W. Powell, P.J. DiMaggio (eds.), Chicago: Univ. of Chicago press, 1991, pp.1-38. [14] V. M. Eguluz, M. G. Zimmerman, C. J. Cela-Conde, M.S. Miguel, Cooperation and the Emergence of Role Differentiation in the Dynamics of Social Networks, American Journal of Sociology, 4 (110), 2005, pp. 977-1008. doi: 10.1086/428716 [15] S. Steinmo, What is Historical Institutionalism? in Approaches and Methodologies in the Social Sciences A Pluralist Perspective, Porta D. (eds). M. Keating. New York: Cambridge University Press 2008. doi:10.1017/CBO9780511801938.008 IMS-2021. International Conference “Internet and Modern Society” 333 [16] V.N. Burkov, A.Yu. Zalozhnev, D.A. Novikov Graph theory in the management of organizational systems. Moscow 2001 (In Rus.) [17] A.A. Azarov, E.V. Brodovskaya, A.Yu. Bubnov, et al, Predictor Mining: Application of Data Mining Methods in Social Computing Problems, Proceedings of SPIIRAS, 3(26), 2013. Pp. 136- 16 (In Rus.) [18] E.V. Brodovskaya, A.Yu. Dombrovskaya, D.N. Karzubov, et al. Developing methodology for “smart” search for political process markers in social media. Monitoring of Public Opinion: Economic and Social Changes, 5, 2017, pp. 79-104. doi: 10.14515/monitoring.2017.5.06 (In Rus.) [19] A.S. Akhremenko, K.S. Stukal, A.P. Petrov, Web or text? Factors of Protest Spread in Social Media: Theory and Data Analysis, Polis. Political Studies, 2, 2020, pp. 73 – 91. doi: 10.17976/jpps/2020.02.06 (In Rus.)