Global Contentious Politics DatasetKoç University, 2022
Global Contentious Politics Dataset (GLOCON) is the first comparative protest event database for the Global South, focusing on Argentina, Brazil, India, Turkey, and South Africa. Applying artificial intelligence methods to local news sources, GLOCON provides and visualizes a number of contentious political level characteristics: event type, (e.g strikes, rallies, boycotts, protests, riots, and demonstrations, i.e. the “repertoire of contention”), participant type (student, worker, militant, etc), organizer type (union, political party, armed organization, NGO etc), ethnicity, urban/rural. GLOCON is one of the major results of a broader project, called Emerging Welfare (EMW). Funded by European Research Council (ERC) Starting Grant, EMW has been led by Assoc. Prof. Dr. Erdem Yoruk, from Koc University, since January 2017.Here, we employ advanced computing techniques such as artificial intelligence, natural language processing, and machine learning in order to extract protest data from online news sources. The GLOCON team has developed fully automated tools for document classification, sentence classification, and detailed protest event information extraction that perform in a multi-source, multi-context protest event setting with consistent performances of recall and precision for each country context. GLOCON's visualization allows the user to observe the distribution of detailed event characteristics over time and space.