The untapped potential of mining news media events for understanding environmental change

Abstract

• Computational social science provides a systematic way of studying online interactions and traditional news media and text. • Social data generated from news media sources can provide near-real-time, geolocated information without language barriers. • Combining news media with social and biophysical data is important to verify results and limit biases in analysis. • Mainstreaming news media mining methodologies enables understanding of social movements and early warning systems for crises. Climate change affects the lives of millions of people. While much attention has been paid to the biophysical impacts of climate change, researchers have little empirical information on the impacts on human society. Climate related social challenges are difficult to accurately measure. One recent data source, the Global Database of Events, Language, and Tone (GDELT) Project, could close that gap. It monitors the world’s broadcast, print, and web news in over 100 languages and identifies the people, locations, organizations, themes, and events driving our global society. Increasingly, big data sources like GDELT are being used to understand how changing actors, events and sentiment in the news media can help understand social change. By analyzing GDELT’s data, applications, and methods, this review identifies the potential of this new data source for the increasingly important role that computational social science can play alongside established biophysical data in monitoring largescale environmental change.