Anzeige
Erschienen in:
2016 | OriginalPaper | Buchkapitel
Identifying Political “hot” Spots Through Massive Media Data Analysis
verfasst von : Peng Fang, Jianbo Gao, Fangli Fan, Luhai Yang
Erschienen in: Social, Cultural, and Behavioral Modeling
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Political processes generated by daily events around the world are highly complicated. Using an information theoretic approach, we examine the dissimilarity of political activities of a country to different countries, and whether it is possible to identify politically “hot” countries. Using a massive political science data, the Global Database of Events, Location, and Tone (GDELT), which covers all the political event data (over 300 million) since 1979 created for studying world-wide political conflict and instability, we show that Shannon entropy for most countries does not differ substantially. Therefore, most countries have similar complexity in terms of political activities. More interestingly, we find that the relative entropies between politically very unstable countries and regular stable countries are large, and thus relative entropy can be used to effectively identify political “hot” spots.