[Asis-l] gdelt 2.0 - a global index over 65 languages and 2, 300 emotions and themes

[kalev leetaru]

Apologies for cross-posting.  I thought many of you would find of great
interest the release yesterday afternoon of GDELT 2.0.  The GDELT Project (
http://gdeltproject.org/) is essentially a live index over the global news
media - within 15 minutes of monitoring a breaking story anywhere in the
world, GDELT is designed to translate it and identify events, counts,
quotes, people, organizations, locations, themes, emotions, relevant
imagery, video, and embedded social media posts, place them into global
context, and make all of this codified metadata available via a live open
metadata firehose enabling open research over global society.

GDELT now updates every 15 minutes, making it possible to track events and
narratives across the global media in near-realtime, along with a large
number of new capabilities of possible interest to many on this list.

Likely of greatest interest is GDELT Translingual, which represents one of
the largest deployments of streaming machine translation, live-translating
what GDELT monitors globally in 65 languages in realtime, representing
98.4% of its daily non-English monitoring volume.  This is coupled with the
first phase of an extremely high-resolution global inventory of the
non-English and non-Western media world to allow GDELT to reach deeply into
local events, reaction, discourse, and emotions of the global media
(especially the non-English and non-Western media) in realtime.

For those interested in geography, there has been a special emphasis on
toponymic recovery across languages, meaning that you can use GDELT 2.0 as
a powerful massively multilingual geocoding system for geofencing media
discussion of particular locations or regions of interest.

In addition, GDELT 2.0 now brings together 24 emotional and thematic
packages to assess a combined 2,300 emotions and themes from every article,
creating an incredible platform to explore sentiment and thematic analysis
at a global scale and in realtime.

Thought this would be of great interest to those on this list trying to
reach across languages, tracking events and narratives across linguistic
and geographic boundaries, and/or wanting to explore the heartbeat of the
global media system in near-realtime.

You can see more technical details below.  I'm working on a set of
tutorials and tools that should be debuting in the next two weeks to make
it much easier to work with the data and not require as much technical
expertise to use it to its fullest extent.

http://blog.gdeltproject.org/gdelt-2-0-our-global-world-in-realtime/
http://blog.gdeltproject.org/gdelt-translingual-translating-the-planet/

For those of you who haven't had a chance to explore GDELT in as much
detail yet, this video gives a very nice overview of the project as a whole:

https://www.youtube.com/watch?v=Psp7YivWL90

Please reach out to me with any questions, I'm very excited to see what
kinds of research are possible when one is able to look across worldwide
media in 65 languages and 2,300 emotions and themes in near-realtime!


~Kalev
http://kalevleetaru.com/
http://blog.gdeltproject.org/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.asis.org/pipermail/asis-l/attachments/20150220/35195638/attachment.html>