Exploring the dynamics of protest with automated computational tools. A Greek case study. (original) (raw)

Protest Event Analysis: A Longitudinal Analysis for Greece

2020

The advent of Big Data has shifted social science research towards computational methods. The volume of data that is nowadays available has brought a radical change in traditional approaches due to the cost and effort needed for processing. Knowledge extraction from heterogeneous and ample data is not an easy task to tackle. Thus, interdisciplinary approaches are necessary, combining experts of both social and computer science. This paper aims to present a work in the context of protest analysis, which falls into the scope of Computational Social Science. More specifically, the contribution of this work is to describe a Computational Social Science methodology for Event Analysis. The presented methodology is generic in the sense that it can be applied in every event typology and moreover, it is innovative and suitable for interdisciplinary tasks as it incorporates the human-in-the-loop. Additionally, a case study is presented concerning Protest Analysis in Greece over the last two d...

Towards Building a Political Protest Database to Explain Changes in the Welfare State

Language Technology for Cultural Heritage, Social Sciences, and Humanities 2016

Despite considerable theoretical work in social sciences, ready to use resources are very limited compared to digitally available mass media resources. Thus, this project creates a political protest database from online news resources in Brazil that will be used to explain Brazilian welfare state policy changes. In this paper we present the preliminary results of a system that automatically crawls digital resources and produces a protest database, which includes events such as strikes, rallies, boycotts, protests, and riots, as well as their attributes such as location, participants, and ideology.

“Just the facts” with PALOMAR: Detecting protest events in media outlets and Twitter

2016

The volume and velocity of available online sources have changed journalistic research in terms of cost and effort re-quired for discovering article topics. However, adding varie-ty and veracity that characterize Big Data, processing it for gathering and organizing information, is not an easy task to handle. The purpose of this study is threefold. Firstly, we present a platform for automated data processing in the con-text of Computational Journalism. We then propose a gen-eral methodology for event extraction from different data sources. Finally, we conducted a pilot implementation of our methodology for protest events extraction from news and Twitter data. Evaluation showed promising results, in-dicating the feasibility of our approach.

Political Claims Analysis: Integrating Protest Event and Political Discourse Approaches

Mobilization

Starting from a critique of protest event and political discourse newspaper analysis, we propose several extensions for a methodological approach which has the quantitative rigor of event analysis, but also retrieves the qualitative discursive elements of claims. Our political claims approach, extends the sample of contentious actions from those of protest event analysis, by coding institutional and civil society actors, and conventional and discursive action forms, in addition to protests by movement actors. This redefines the research object to acts of political claims-making in a multi-organizational field. We use examples from a research project on mobilization on migration and ethnic relations in Britain and Germany, to demonstrate the extra analytic gains that are possible from our approach. By situating protest and social movements not just theoretically, but also methodologically in a wider context of political claims-making, we are in a better position to follow the recent calls for more integrated approaches, which place protest within multiorganizational fields, link it to political opportunities and outcomes, and are sensitive to its discursive message.

Capturing Planned Protests from Open Source Indicators

AI Magazine

Civil unrest events (protests, strikes, and “occupy” events) are common occurrences in both democracies and authoritarian regimes. The study of civil unrest is a key topic for political scientists as it helps capture an important mechanism by which citizenry express themselves. In countries where civil unrest is lawful, qualitative analysis has revealed that more than 75 percent of the protests are planned, organized, or announced in advance; therefore detecting references to future planned events in relevant news and social media is a direct way to develop a protest forecasting system. We report on a system for doing that in this article. It uses a combination of keyphrase learning to identify what to look for, probabilistic soft logic to reason about location occurrences in extracted results, and time normalization to resolve future time mentions. We illustrate the application of our system to 10 countries in Latin America: Argentina, Brazil, Chile, Colombia, Ecuador, El Salvador,...

A Task Set Proposal for Automatic Protest Information Collection Across Multiple Countries

European Conference on Information Retrieval, 2019

We propose a coherent set of tasks for protest information collection in the context of generalizable natural language processing. The tasks are news article classification, event sentence detection, and event extraction. Having tools for collecting event information from data produced in multiple countries enables comparative sociology and politics studies. We have annotated news articles in English from a source and a target country in order to be able to measure the performance of the tools developed using data from one country on data from a different country. Our preliminary experiments have shown that the performance of the tools developed using English texts from India drops to a level that are not usable when they are applied on English texts from China. We think our setting addresses the challenge of building generalizable NLP tools that perform well independent of the source of the text and will accelerate progress in line of developing generalizable NLP systems.

Protest Event Dataset for Croatia, Portugal, Serbia and Spain

Politička misao, 2021

The protest event analysis (PEA) dataset covers the period 2000-2017, recording protest events in Spain, Portugal, Croatia and Serbia. Best described as "a type of content analysis" (Hutter, 2014: 335), PEA systematically captures properties of protest events such as frequency, timing, duration, location, claims, size and others (Koopmans and Rucht, 2002). Our dataset of protest events is based on domestic daily newspapers' reports in each of the four countries. Notwithstanding selection and description bias inherent to media reporting (for reviews, see

Mapping protest on the refugee crisis: insights from online protest event analysis

This chapter chapter draws on a protest events analysis conducted through Google News to detect the contentious dynamics triggered by refugee crisis in Greece, Spain and Italy. Analysis shows differences and similarities in the protest fields in the three countries mainly due to the level of exposition to migration trajectories filtered by domestic political opportunities: if in Spain, a very peculiar geo-political situation minimizing the migration trajectories resulted in few and basically pro-immigrants oriented protests; in Italy, the exposition of the country to the migration trajectories during the refugee crisis and the politicization of the issue by right-wing parties and organizations, combined to shape a highly contentious interaction between pro and anti immigration protest actors, both manly targeting domestic institutions. The Greek protest field is found in between these two opposite situations, but a bit closer to Spain: on the one hand, the anti-immigration frame received little support by the protest field actors, on the other hand, EU institutions were the main target of the protests, especially after the EU-Turkey agreement.

Big Data, Social Media, and Protest: Foundations for a Research Agenda

Computational Social Science

Following the Arab Spring, a debate broke out among both academics and pundits as to how important social media had been in bringing about what may have been the least anticipated political development of the 21 st century. Critics of the importance of social media pointed in particular to two factors: (a) the proportion of social media messages that were transmitted in English; and (b) the proportion of Arab Spring related social media posts that originated from outside the Arab world. In our chapter, we will test whether two important subsequent unanticipated protests-Turkey's 2013 Gezi Park protests and Ukraine's 2013-14 Euromaidan protest-also are susceptible to such criticisms. To do so, we draw on millions of tweets from both protestsincluding millions of geolocated tweets from Turkey-to test hypotheses related to the use of Twitter by protest participants and, more generally, in-country supporters of protest movements, effectively refuting the idea that emerged after the Arab Spring that Twitter use during protests only reflects international attention to an event.

Mapping Movements - Social Movement Research and Big Data: Critiques and Alternatives

The integration of social media and other online and mobile platforms and technologies into social movements around the globe has received significant academic attention around questions such as 'did social media cause a successful revolution?' and 'what information flows are apparent during protests on social media?'. Phenomena such as the Occupy Movement and the Arab Spring, which both saw widespread social media activity, have been the subjects of numerous studies in different fields, taking advantage of the wealth of social media data available (for example, Conover, Davis, et al., 2013; Costanza-Chock, 2012; Gaby & Caren, 2012; Papacharissi & Oliveira, 2012; Starbird & Palen, 2012). Automated processes and tools for capturing and processing such activity have meant that large social media datasets, of tweets in particular, are increasingly common sources for research into social movements and their activities and coverage online. However, the growing use of big data, social media-oriented approaches in the study of social movements raises new analytical and ethical challenges. There are important differences between big data research methodologies and previous approaches to social movement research, including a radically altered relationship between researchers and movement participants. Social media data capture and analysis around these topics can be carried out without having to be physically near or involved in the movements in question, which raises concerns about how to evaluate potential risks to participants; reciprocity; the accessibility of research to activists, including for comment and criticism; and how researchers engage with movement participants as knowledge-producers. Analytically, the use of big data methods for social movement research requires a careful attention to the biases in available data. Biases and gaps in the data may be introduced through strategic avoidance of social media or self-censorship by activists; the limitations of platform architecture and content policies; practices such as subtweeting and screen-capping which deliberately obscure links between accounts; the use of images and other non-text forms not captured by big data tools; the openness of different social media platforms to data capture; and the limitations of data capture tools themselves. In response to these challenges and concerns, we advocate the use of a mixed-methods approach that combines participant observation, in-depth interviews, and big data methods. This approach offers a framework for balancing the benefits of new quantitative methods with a need to prioritise an ethical approach to social movement research, as well as correcting some of the biases introduced by big data methods. This approach has been developed through the Mapping Movements project, which has examined movements and events in North America, Africa, and Europe. This chapter draws most prominently on the first published case study of the project, looking at the use of Twitter within the Occupy Oakland movement (Croeser & Highfield, 2014). This research demonstrates that a mixed-methods approach allows a better understanding of the contexts of social movements and their uses of social media. Considering both the online and the physical aspects of social movements enables a nuanced analysis of social media use by activists, looking beyond the object of study (the social medium of choice) at a quantitative level, to examine the intersections between these aspects of social movements. Crucially, our work demonstrates how blended methods can combine the strengths of different research approaches to collectively overcome the limitations of big, social media data, providing detail and explanation for activity found in – and hidden from – these datasets, addressing some of the gaps in big data research. We have not, however, dealt with the Occupy Oakland case study in detail here; rather, we have attempted to outline some of the most pressing ethical and analytical issues for big data research on social movements which are more broadly relevant, and to offer potential avenues which may provide (partial) solutions.