What is GDELT Project?
The GDELT Project monitors print, broadcast and web news from nearly every country in over 100 languages and turns it into open data about the people, organisations, locations, themes, emotions, quotes, images and events in the coverage. Supported by Google Jigsaw, it updates every 15 minutes and its event archive goes back to January 1, 1979. News in 65 languages is machine translated into English in real time so it can be processed together.
Researchers can query the data in several ways: the Event Database and Global Knowledge Graph as raw files or in Google BigQuery, live JSON APIs for news articles (DOC), geography (GEO) and television (TV), and the browser-based GDELT Summary. Analysts and journalists use it to track coverage of conflicts, protests and other events across countries and languages, and to see how a story spreads through the world's media.
Main Features
- Monitoring of world news in over 100 languages, updated every 15 minutes
- Event Database with records back to 1979
- Global Knowledge Graph of people, organisations, places, themes and emotions
- Real-time translation of news in 65 languages into English
- DOC, GEO and TV APIs returning live JSON
- Full datasets in Google BigQuery and as raw CSV downloads
- Specialised datasets such as the Global Frontpage Graph and Global Quotation Graph
How to Get Started with GDELT Project
- Go to gdeltproject.org and open the Data page to see the available datasets.
- Try GDELT Summary or the DOC API to search recent news coverage on a topic without writing code.
- Use the GEO API or the Event Database to map where events are being reported.
- For large analyses, query the tables in Google BigQuery or download the raw files.
GDELT Project Pricing
GDELT says its entire database is 100% free and open. Querying it in Google BigQuery is subject to Google Cloud's own usage charges beyond BigQuery's free allowance.
Who Should Use It?
- Researchers studying global events, conflict and protest
- Journalists tracking how a story is covered around the world
- Data scientists who need large multilingual news datasets
- Analysts monitoring news about a country, company or issue
- Students learning computational social science
GDELT Project Alternatives
- OpenSanctions: a free, open database of sanctioned people, companies and politically exposed persons, built from hundreds of official sources.
- Social Searcher: a social media search engine for tracking mentions across networks.
- Wayback Machine: the Internet Archive's record of how web pages looked in the past.
Frequently Asked Questions
Is GDELT Project free?
Yes. GDELT says its entire database is free and open. The APIs and raw downloads cost nothing, while large queries in Google BigQuery can incur Google Cloud charges.
How far back does GDELT go?
The Event Database goes back to January 1, 1979, and the newer GDELT 2.0 datasets update every 15 minutes. Some specialised datasets cover shorter periods, which are listed on the Data page.
Is GDELT data accurate?
GDELT codes events automatically from news reports, so it reflects what the media reported rather than verified facts, and the same event can appear many times. Treat it as a guide to coverage and trends, and check important findings against the original articles.