Analysis of Social Response to Incidents: A Comparative Study of Arkhangelsk and Yekaterinburg

The Aim of this study is to develop and validate a methodology for analyzing social media data to enhance the reliability of emergency response systems by enabling rapid identification, assessment of scale, location, and forecasting of crisis events based on regional specifics of public activity. To achieve this goal, a Python‑based program was developed to automate the collection, preprocessing, and comprehensive analysis of data from the VKontakte social network. The program was tested using data from two cities with different demographic characteristics and socio‑economic conditions, i.e., Arkhangelsk and Yekaterinburg.

Methods. The research employed methods of data collection from open VKontakte groups, followed by preprocessing and comprehensive analysis. The analysis included text analysis (sentiment analysis and word cloud generation), mathematical analysis (entropy and entropy derivative calculations to assess activity dynamics), and external factor analysis (the influence of meteorological conditions, holidays, and weekends).

Results. The study revealed significant regional differences in social activity levels across various categories of emergency situations. Activity levels in Arkhangelsk were at least twice as high as those in Yekaterinburg, despite the smaller population of the city. The nature of activity also differed significantly: sharp spikes in activity were observed in Yekaterinburg, while activity in Arkhangelsk was more evenly distributed. Seasonality manifested in increased activity during periods of technical work or extreme weather conditions. In the “Fire” category, both cities demonstrated high and sustained activity; however, sharper spikes were noted in Yekaterinburg, particularly at the end of March and beginning of April, potentially indicating major incidents. In the “Water Outage” category, two significant peaks in activity were recorded in Arkhangelsk, in April and onJuly 31 and August 1, possibly pointing to widespread water supply issues. In Yekaterinburg, activity in this category was lower but more frequent, likely reflecting minor disruptions or informational updates.

Conclusion. Social networks serve as a valuable source of data for analyzing public reactions to emergency situations. The identified regional characteristics of user behavior high light the need to create adaptive monitoring and forecasting systems that account for the specific features of each region. Using data from social networks enhances the reliability and efficiency of response systems by enabling rapid determination of incident scale, location, and consequences, as well as identifying seasonal and local threats. The findings confirm the necessity of implementing automated analytical tools capable of promptly assessing situations. Social networks can act as indicators of seasonal and local threats, allowing for proactive risk preparation. The observed differences in activity levels between regions underscore the importance of considering local conditions when developing strategies for communication and crisis management. Active use of social networks as a platform for civic participation demonstrates their potential to strengthen interaction between the public and authorities during crises.

1. Timashev S., Malyutina E. Automated entropy analysis of the social consequences of urban man-made accidents and natural catastrophes. Reliability: Theory & Applications 2024;19:SI6(81):142-150.

2. Timashev S.A., Malyutina E.V. [Entropic analysis of social activity during and after major accidents of urban infrastructure]. Yekaterinburg: Uralsky Rabochy; 2023. (in Russ.)

3. Timashev S., Malyutina E. Entropy Analysis of Social Unrest After Large Urban Infrastructure Accidents. ASCE INSPIRE 2023;1:631–639. DOI: https://doi.org/10.1061/9780784485163.074.

4. Diakopoulos N., De Choudhury M., Naaman M. Finding and assessing social media information sources in the context of journalism. In: Proceedings of the SIGCHI conference on human factors in computing systems; 2012. Pp. 2451-2460. DOI: 10.1145/2207676.2208409.

5. Kwak H. et al. What is Twitter, a social network or a news media? In: Proceedings of the 19th international conference on World wide web; 2010. Pp. 591-600. DOI: 10.1145/1772690.1772751.

6. Starbird K., Muzny G., Palen L. Learning from the crowd: Collaborative filtering techniques for identifying on-the-ground Twitterers during mass disruptions. ISCRAM; 2012.

7. Zubiaga A., Ji H., Knight K. Curating and contextualizing twitter stories to assist with social newsgathering. In: Proceedings of the 2013 international conference on Intelligent user interfaces; 2013. Pp. 213-224. DOI: 10.1145/2449396.2449424.

8. Gao Y. et al. Brand data gathering from live social media streams. In: Proceedings of international conference on multimedia retrieval; 2014. Pp. 169-176. DOI: 10.1145/2578726.2578748.

9. Olteanu A. et al. Social data: Biases, methodological pitfalls, and ethical boundaries. Frontiers in big data 2019;2:13. DOI: 10.3389/fdata.2019.00013.

10. Sung K.H., Lee M.J. Do online comments influence the public’s attitudes toward an organization? Effects of online comments based on individuals’ prior attitudes. The Journal of psychology 2015;149(4):325-338. DOI: 10.1080/00223980.2013.879847.

11. Hasan M., Orgun M.A., Schwitter R. Real-time event detection from the Twitter data stream using the TwitterNews+ Framework. Information Processing & Management 2019;56(3):1146-1165. DOI: 10.1016/j.ipm.2018.03.001.

12. Teodorescu H.N. Using analytics and social media for monitoring and mitigation of social disasters. Procedia Engineering 2015;107:325-334. DOI: 10.1016/j.proeng.2015.06.088.