Online News Summary
Below is a visual dashboard summary of global online news media coverage of your search, updated to 15 minutes ago. Refresh this page again in 15 minutes to see the most recent updates. Displays work best in Google Chrome.
More Info.
All visualizations are powered by the GDELT Project's DOC 2.0 and GEO 2.0 APIs - technical users will find "export" options on each of the visualizations on this page that allows them to download the underlying CSV/JSON/GeoJSON data for further visualization/analysis and to view the HTML code that allows them to embed each individual visualization on their own web page - the embedded visualization will update each time it is viewed/refreshed to display the most recent data as of 15 minutes prior.
Search Details
The displays on this page summarize coverage returned by the following search:
Data Source: Global Online News Coverage
Human Summary: theme:env_biofuel AND PublicationDate<=1 Year
[Modify Search]
Volume Timeline
The timeline below shows the percent of all global online news coverage monitored by GDELT over the given day/hour that matched your query, allowing you to trace how attention to your topic has changed over time and whether it is increasing or decreasing. (Note that even if you have limited your search to a particular country or language, the results are still reported as a percent of all online media monitored by GDELT, not just the media from that country/language, so in practice you should focus on how the timeline changes over time, rather than the precise percentage at any given point in time).
Move your mouse over the timeline to see the value for a given day/hour. You can also zoom into a particular time period by clicking on the timeline and dragging to the right or left to select that date/time range. The entire GDELT Summary display will then refresh to reflect just coverage from that zoomed-in time period (though maps will be unchanged and still show the last 24 hours of data). In this way, if you see a large peak in coverage on a particular day, you can zoom in and see what that coverage focused on and where it came from.
Click on the Export button at the top right of the visualization to save it as an image, download the underlying data as CSV/JSON for further statistical analysis or get the code to embed it on your own web page.
Tone Barchart
The tone barchart below is a unique display that takes all matching coverage monitored by GDELT over the selected time period, computes the document-level tone of each article (ie, the tone of the article as a whole rather than the tone just refering to your keywords) and bins them into a histogram/barchart that ranges from extremely negative on the left to extremely positive on the right.
Traditional sentiment analysis tools tend to report average tone as a single number, such as "The average tone of Putin-related articles is '-4'". Reducing millions of matching articles to a single number isn't very helpful, as that doesn't tell you whether most of that Putin-related coverage hovered around -4 or whether coverage was largely stratified at extremely negative and extremely positive and -4 just happened to be the average of those scores. Instead, this tone barchart display shows you the actual histogram distribution so you can see where most of the coverage is clustered.
You can mouse over each bar to see a few of the top most relevant matching articles that were scored with that tone level, letting you instantly skim coverage from very negative to very positive. Clicking any link will open it in a new browser window. Note that tone is calculated at the level of the entire document, rather than just the tone relating to your keyword - hence a search for a given person/organization will return the average tone of the entirety of the articles mentioning that person, rather than the tone of just the sentences mentioning that person/organization.
Click on the Export button at the top right of the visualization to save it as an image, download the underlying data as CSV/JSON for further statistical analysis or get the code to embed it on your own web page.
Locations Map: Countries
GDELT automatically identifies all geographic locations mentioned in each article and can create a map showing the top thousand or so worldwide locations that are mentioned most frequently within a few sentences of your keyword(s) over the last 24 hours. These algorithms are 100% automated, so you may see errors in this map, but it offers a useful overview of what locations are most associated with your keywords over the last 24 hours. All locations are aggregated to the country level. You can click on a country to see a few matching articles that mention your keyword(s) in the context of locations in that country. NOTE that the map below shows data from the past 24 hours, regardless of any date filtering you applied.
Click on the "GeoJSON" link at the bottom right of the map to download the map data in GeoJSON format, suitable for import into any mapping or geospatial platform for further visualization and analysis.
Top Articles
Below is a short list of some of the most relevant coverage that matched your search. If the article included an editorially-selected display image (the news outlet embedded special HTML code in the article declaring an image to show when sharing or linking to the article), it will be displayed beside it. Note that the images shown are those editorially selected by the publisher of each article and do not reflect any image-specific search terms you applied. Clicking any link will open it in a new browser window. If a URL is no longer accessible, you can try viewing it through the Internet Archive's Wayback Machine web archive.
Top Images
Below is a short list of some of the most relevant images that matched your search. If your search includes image-specific search options or filters, the most relevant images will be selected, otherwise if your search contains only textual search options, the first set of images from the most relevant articles matching your search will be shown. Unlike the "Top Matching Articles" "Visual List" mode, which simply displays the "social sharing image" of each article, the images below were selected based on GDELT's processing of a random subset of up to one million global news images per day through Google's Cloud Vision API deep learning algorithms, meaning they have actual semantic bearing on your search and if you include image-related terms/filters in your search, you can perform deep learning-powered image search. Note that not every image from every article is able to be processed each day and a news outlet may remove an image after publication, so the images shown below do not reflect an exhaustive inventory of 100% of all global news imagery. On a slow internet connection it may take awhile for the images below to fully load. Clicking any link will open it in a new browser window. If a URL is no longer accessible, you can try viewing it through the Internet Archive's Wayback Machine web archive.