New Dataset Collects Instances of ‘Contentious Politics’ Around the World
In a world where conflict, turmoil, and polarization have become societal norms, it is increasingly important for scholars and policymakers to be plugged into the happenings around the globe.
To effectively provide access to this information, the European Research Center is funding the Global Contentious Politics Dataset, or GLOCON, a state-of-the-art automated database curating information on political events — including confrontations, political turbulence, strikes, rallies, and protests – in Brazil, South Africa, Argentina, India, and Turkey. The project website notes that expected updates will “incorporate many other countries from all around the world and aim at producing a truly global contentious politics database.”
The database is hosted at the Center for Computational Society Sciences at Istanbul’s Koç University.
GLOCON uses machine learning, artificial intelligence, and natural language processing to collect and analyze its current collection of 621,290 different instances of contentious politics as drawn from various local news sources.
In a message from sociologist Erdem Yörük, director of the Center for Computational Social Science, “GLOCON was originally designed [during the Emerging Markets Welfare Project] to examine the interplay between contentious politics and social welfare schemes in the Global South, but we hope the dataset will be useful for a broader academic community, including social movement, conflict and computational social science scholars.”
Data on the website ranges from the 1990s to current-day events and will include detailed specifics on location, time, organizers, and participants of contentious political events. The database will also provide an interactive dashboard, distinguishing between violent vs. non-violent and rural vs. urban events, particularly in countries like South Africa and India.
The tool offers data visualizations drawn from categorical, geolocational, and temporal filters, and users can gain access to raw data via the download feature. An annotation manual provides a detailed view of the database’s methodology and training data.
