Early warning signals for war in the news

Bibliographic Data

ID3916631
AuthorsThomas Chadefaux (0000-0002-8456-8124, ETH Zurich, corresponding author)
Year2014
Volume51
Issue1
Pages5-18
Publication date2014-01-01
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueJournal of Peace Research (JOURNAL)
Journal identifiersISSN: 0022-3433 • E-ISSN: 1460-3578
PublisherSAGE Publications Inc (PUBLISHER)
DOI10.1177/0022343313507302
OpenAlexW3122027301
LanguageEN
Citations received38
References cited51

There have been more than 200 wars since the start of the 20th century, leading to about 35 million battle deaths. However, efforts at forecasting conflicts have so far performed poorly for lack of fine-grained and comprehensive measures of geopolitical tensions. In this article, a weekly risk index is derived by analyzing a comprehensive dataset of historical newspaper articles over the past century. News reports have the advantage of conveying information about contemporaries' interpretation of events and not having to rely on meaning inferred a posteriori with the benefit of hindsight. I applied this new index to a dataset of all wars within and between countries recorded since 1900, and found that the number of conflict-related news items increases dramatically prior to the onset of conflict. Using only information available at the time, the onset of a war within the next few months could be predicted with up to 85% confidence and predictions significantly improved upon existing methods both in terms of binary predictions (as measured by the area under the curve) and calibration (measured by the Brier score). Predictions also extend well before the onset of war - more than one year prior to interstate wars, and six months prior to civil wars - giving policymakers significant additional warning time

Ancient history · Battle · Brier score · China · Geopolitics · Hindsight bias · Index (typography) · Meaning (existential) · Newspaper · Political science · Politics · Spanish Civil War · Artificial Intelligence · Climate variability and models · Computer Science · Data-Driven Disease Surveillance · History · Law · Psychology · Social Psychology · Yersinia bacterium, plague, ectoparasites research

Tools

Open DOISci-Hub