Emigration narratives relevance assessment by large language models for social media monitoring
Download paperDoctor of Sociology, Chief Researcher, Center for Political Studies, Financial University under the Government of the Russian Federation, Moscow, Russia; ORCID 0000-0002-2015-2349
e-mail: an-doc@yandex.ruDoctor of Political Sciences, Chief Researcher, Center for Political Studies, Financial University under the Government of the Russian Federation, Moscow, Russia; ORCID 0000-0001-5549-8107
e-mail: brodovskaya@inbox.ruSection: Artificial Intelligence in Media and Communication Studies
The paper’s relevance lies in the lack of methodological experiments specifically focused on evaluating the heuristics of artificial intelligence (AI) for assessing users’ social sentiments expressed through digital markers, particularly in the context of Russian national audience’s emigration intentions. The article delves into studies that measure language and text as data, with a special focus on understanding users’ social attitudes and behaviors expressed through digital markers. The primary objective of the study is to assess the significance of the streams downloaded by the neural network, specifically the LSTM language model employed by the Medialogia service. The research design encompasses cognitive mapping (a preliminary stage to identify search queries) and social media analysis conducted by the Medialogia service. Additionally, a representative sample of the downloaded streams is manually analyzed to evaluate their relevance. The paper highlights common errors in creating search queries and provides strategies to overcome these inaccuracies, which can be further utilized to enhance the neural network’s ability to download relevant datasets. Furthermore, the relevance of the streams segmentation conducted by the language model is analyzed. The paper makes an assumption about the underlying reasons for the varying degrees of relevance of documents (posts) downloaded by the service.
DOI: 10.55959/msu.vestnik.journ.5.2025.209232References:
Aroles J., Bonneau C., Bhankaraully S. (2022) Conceptualising ‘Meta-Work’ in the Context of Continuous, Global Mobility: The Case of Digital Nomadism. Work, Employment and Society 37: 1261–1278.
Benoit K. (2020) “Text as Data: An Overview.” In L. Curini, R. Franzese (eds.) The SAGE Handbook of Research Methods in Political Science and International Relations. London: SAGE Publications. Pp. 461–497.
Brodovskaya E. V., Dombrovskaya A. Yu., Pyrma R. V., Azarov A. A. (2020) Informatsionnye potoki o migrantakh i dlya migrantov v sotsial’nykh media Rossii [Information flows about and for migrants in Russian social media]. Informatsionnoe obshchestvo 6: 7–23. (In Russian)
Bronitsky G. (2024) Migration nowcasting using Google Trends: cross-country application. Population and Economics 8 (2):133–154.
Bagdasaryan V. E., Volodenkov S. V., Zhmurin I. E. (2025) Chelovek i tekhnologicheskiy progress: antropologicheskaya povestka mirovogo razvitiya: monografiya [Humankind and technological progress: an anthropological agenda for global development: monograph]. Yaroslavl: Shukaeva i sem’ya Publ. (In Russian)
Cook D. (2023) What is a digital nomad? Definition and taxonomy in the era of mainstream remote work. World Leisure Journal 6: 256–275.
Glushchenko G. I. (2021) Razvitie virtual’noy migratsii v kontekste tsifrovizatsii [The development of virtual migration in the context of digitalization]. DEMIS. Demograficheskie issledovaniya 1 (2): 57–64. (In Russian)
Grebenyuk A. A., Subbotin A. A. (2021) Issledovanie migratsionnykh protsessov v elektronnukh sotsial’nykh setyakh [Research on migration processes in online social networks]. Tsifrovaya sotsiologiya / DigitalSociology 4 (2): 23–31. (In Russian)
Hochreiter S., Schmidhuber J. (1997) Long Short-Term Memory. Neural Computation 9 (8): 1735–1780.
Iskusnykh O. (2021) Internet kak faktor rasprostraneniya emigratsionnykh ustanovok sredi molodezhi [The Internet as a factor in the spread of emigration attitudes among youth]. Kognitivnye nauki v informatsionnom obshchestve1 (2). (In Russian)
Le Mens, G. and Gallego, A. (2025) Positioning Political Texts with Large Language Models by Asking and Averaging. Political Analysis 33 (3): 274–282. DOI 10.1017/pan.2024.29
Pu T., Huang Ch., Zhang H., Yang J., Huang M. (2024) Application of deep learning model incorporating domain knowledge in international migration forecasting. Data Technologies and Applications 58 (5): 787–806.
Volkov Yu. G., Krivopuskov V. V., Kurbatov V. I. (2021) Tsifrovye migranty i tsifrovaya diaspora: novye problemy i trendy mezhdunarodnoy migratsii [Digital migrants and the digital diaspora: new problems and trends in international migration]. Tsifrovaya sotsiologiya / DigitalSociology 4 (4): 102–108. (In Russian)
Volodenkov S. V. (2017) Total Data kak fenomen formirovaniya politicheskoy postreal’nosti [Total Data as a phenomenon of political post-reality formation]. Vestnik Omskogo universiteta. Seriya “Istoricheskie nauki” 3 (15): 409–415. (In Russian)
Welch N. G., Raftery A. E. (2022) Probabilistic forecasts of international bilateral migration flows. Proceedings of the National Academy of Sciences of the United States of America 119 (35): 1–8.
To cite this article: Dombrovskaya A. Yu., Brodovskaya E. V. (2025) Otsenka relevantnosti identifikatsii emigratsionnykh tekstov bol’shimi yazykovymi modelyami dlya monitoringa sotsial’nykh media [Emigration narratives relevance assessment by large language models for social media monitoring]. Vestnik Moskovskogo Universiteta. Seriya 10. Zhurnalistika 5: 209–232. DOI: 10.55959/msu.vestnik.journ.5.2025.209232

