The introduction
In the era of new technologies, the number of new gadgets and various social networks is increasing daily. Few people think, but based on an understanding of the dynamics of protests, one can track possible cases of their appearance using data from social networks. Just as social networks influence the development of mass protests, mass protests leave their mark on social networks, that is, posts or personal messages. The purpose of this paper is to summarize the main issues of the article, analyze the results, and identify the problem.
Summary of the core issues
The presented article is about a new and relevant topic in our time, which involves determining the geo-position of possible mass protests using social networks. Wilson (2017) finds “strong evidence of mass protest in a state being associated with an increase in social media activity nationwide” (p.7). Thus, there is a decrease in any social activity in the venue of the protest, but at the same time, its scale is increasing. The purpose of the article is to study how social networks geospatially respond to protests using Twitter data.
Research of Twitter data obtained before, during, and after the events of Euromaidan in Ukraine in 2014 are used as an example. The author also reflects on the impact of social networks on mass protests and gives two concepts of this influence. The article explores the approach to geocoding posts on Twitter, which does not depend on the availability of media or the content in it. Besides, this approach studies how important the user is for the entire social network and how active it is in the network. Thus, more engaged users have a more considerable influence on the occurrence of mass protests.
The problem
A protest is a strong objection to something, and a mass demonstration is a group objection. Often such events cause negative consequences; moreover, they can cost someone’s life. Over time, it has become much easier to find out about such protests with the help of the media (Lee & Chan, 2010). However, most often, they describe the process rather than trying to prevent it. The author of the article cites theories that explain the relationship between protests and social networks or the media.
Years earlier, this task was accomplished by manually coding newspaper newsletters from regions and cities of the country or by the census of specific people who had previously attended such protests. The problem of measuring the dynamics of cases of mass protests is a long-standing one in the social sciences (Wilson, 2017). However, over the past few years, more new technologies have appeared that simplify the encoding of information. Mainly due to the latest software for recognizing faces and images of people who may be involved in holding mass protests. Computers analyze new posts by keywords, for example, “violence” and use contextualization technology; that is, they determine the location, gender, and age of the user, the number of “likes” on this post, etc.
This approach has several limitations since it entirely depends on the availability of electronic versions of geographic locations that are of most interest to users. The author claims that authoritarian states that control social networks can become an obstacle to this approach since, without their permission, it is impossible to obtain these networks fully (Wilson, 2017). Another problem in the theory of this study is the adaptation of context algorithms to a specific language. This method requires not simple dictionary translations, but a complete understanding of information about possible events in different languages of the world, including slang. In this case, the technique cannot provide reliable information contained in the post.
The article discusses two main theoretical concepts of the impact of social networks and the internet on mass protests. The first hypothesis states that the internet, social networks, and applications are the “organizer” of riots, so they are discussed before the event. The second hypothesis suggests that social networks are exclusively informative; they describe an event that has already occurred. The author claims that in practice, the differences between the two theories are difficult to verify (Wilson, 2017). Both concepts are correct, since the organization of the protest before it can take place, as well as its description after the event, complicates the process of data for research.
Researches
To study the influence of protests on social networks described earlier by the method, two data sets are needed; that is, data from social networks and data about mass protests. The author also points out that it is necessary to study their scale and time frame. An example of such a study is Euromaidan in Ukraine in 2014, and data are collected on its conduct from October 1, 2013, to March 31, 2014, that is, the situation is covered before, during, and after Euromaidan.
Based on the obtained statistical results for every day for six months, it can be concluded that the largest number of encoded channels was from November to February, that is, during the most massive protests on Euromaidan. The author also provides a research map that describes the number and mass of demonstrations in different areas of the city. Thus, people in the periphery used social networks much more than those who were in the city center and directly participated in the mass protest (Wilson, 2017). Users who were in the city center during this period posted much fewer publications.
The author found that the mass protests in Kyiv were associated with a decrease in activity on social networks in the center of the city and its increase on the outskirts (Wilson, 2017). Thus, the existence of this phenomenon can be verified in other cities and countries to find out whether this is an exceptional feature of Euromaidan. A similar study is also being conducted in Belarus and Russia, and in Lviv, the same regularity has already been identified, which may indicate that the method is universal.
Conclusion
To summarize, the article highlighted the main aspects of the impact of mass protests on social networks and the media; also, it studied how the internet geospatially responds to protests using Twitter data. The author gave theoretical hypotheses about the differences of this influence, that is, organizational and describing. The main argument is the ratio of the geo-location of the user who posted the picture or video and the publication time.
He also used the method of content analysis, that is, the search for the posts necessary for research by keywords. Thus, having obtained statistics on the study of Euromaidan, we can conclude that the number of jobs in the place where the protest takes place decreases. This article may be useful to statisticians who study the dynamics of mass protests, as it describes a new method for studying the interdependence of social networks and protests.
References
Wilson, S. (2017). Detecting mass protest through social media. The Journal of Social Media in Society, 6(2), 5-25.
Lee, F. L., & Chan, J. M. (2010). Media, social mobilization and mass protests in post-colonial Hong Kong: The power of a critical event. New York, NY: Routledge.