GDELT was mentioned briefly during the "DATA & ANALYTICS: Pushing the Limits of BigQuery" session at the Google Cloud Platform Global User Conference NEXT 2016. Watch the Full Video.
Kalev's latest piece for Forbes explores the changing landscape of ethics in the big data world. Read the Full Article.
Kalev's latest Forbes piece explores how simply having more data does not automatically lead to better decisions being made. Read the Full Article.
Kalev's latest Forbes piece explores how AI and robotics are taking over human society. Read the Full Article.
Kalev's latest Forbes piece explores the changing landscape of publishing in the digital era. Stanford University's Law Library even excerpted several key passages from the piece, calling it a "though provoking" article. Read the Full Article.
Kalev's latest piece for Forbes explores how technology is turning us into digital zombies and how AI may pull us back from the brink. Read the Full Article.
Kalev's latest piece for Forbes explores whether we could ever truly have a free Internet. Read the Full Article.
Kalev's latest piece for Forbes explores the dangerous gap between the producers and consumers of data sciences today and why we need more domain experts for the field to reach its full potential. Read the Full Article.
Kalev's latest piece for Forbes explores how the commercial cloud has upended and revolutionized how computing is done. Read the Full Article.
Tiffany Trofino, Principle Strategist at IBM, wrote recently on LinkedIn about the future of using data for understanding global conflict. Read The Full Article.
Putting together data from the television tracker, Cloud Vision analysis of political imagery, and Jordan and Felipe's talk from Google I/O 2016, Kalev's latest piece for the Washington Post focuses on the state of the Democratic race and how the data clearly showed Bernie on the wane long before his eventual loss. Read the Full […]
With the debut of the GDELT Visual Global Knowledge Graph (VGKG), which uses Google's Cloud Vision API deep learning algorithms to catalog global news imagery, we've been immensely excited about the ways this incredible technology can be used to catalog and understand global visual narratives. This time, Felipe Hoffa came up with an incredible way of combining the GKG […]
Researchers Aaron Schein (University of Massachusetts Amherst), Mingyuan Zhou (University of Texas at Austin), David Blei (Columbia University) and Hanna Wallach (Microsoft Research) presented "Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations" at the 33rd International Conference on Machine Learning. Read the Full Paper.
One of the amazing things that happens when you monitor the world at GDELT's scale is that you begin to observe language evolving in realtime. New words and grammatical constructs come into being, older words and structures fade from use, punctuation rules change and the topics, contexts and ways in which language use are in constant flux. […]
Ever since 2010's Culturomics paper and with the rise of ever-more powerful linguistic modeling systems, ngrams have surged back into popularity as a quick, but primitive, way of studying language. On the surface, ngrams would appear to be quite simplistic to compute: just split each document into words and count up how many times each […]
In their lively and fast-paced Google I/O 2016 talk, Felipe Hoffa and Jordan Tigani trace mainstream and social media coverage of the 2016 US presidential election through the eyes of GDELT and several other datasets. One particularly striking graph is the timeline below, showing the intensity of mentions of Bernie Sanders in the mainstream media (via […]
Felipe Hoffa and Jordan Tigani offered a fantastic fast-paced look at Google BigQuery in their session at Google I/O 2016, showcasing how GDELT can be used to understand the media dynamics of Election 2016. Check out their talk below or read their session description. Fast forward to around 17:24 to see the television tracker at work and […]
BBVA's latest report, the EAGLEs Economic Outlook Annual Report 2016, has a wide array of quite fascinating sentiment and other analyses based on GDELT and shows how the GDELT Global Knowledge Graph can be used to dive deeply into complex topics. Read the Full Report.
We recently made three bug fixes to address two geographic issues and one person name extraction issue. Mapping the textual geography of the world's news media is an incredibly difficult task and contextual disambiguation plays a critical role in generating robust results. In some cases the geocoding infrastructure makes use of additional external domain knowledge beyond that contained in […]
CuriousGnu used BigQuery put together a set of fantastic maps in CartoDB using GDELT to map the average tone of the media coverage of each country around the world about Hillary Clinton and Donald Trump, producing some fascinating findings! At the most obvious level, the world's media seems to like Trump a lot less than […]
Kalev gave the opening keynote of the second day of Data Summit 2016 in New York City last week, speaking on the power of big data to reshape how we understand the world around us. Learn More. Conference Program. Keynote Writeup.
Kalev's latest Forbes story "Is Facebook's Trending Topics Biased Against Africa And The Middle East?" includes a map of the geographic footprint of the 1,000 media outlets powering Facebook's Trending Topics module, visualizing the tremendous Western bias in the list and the tremendous paucity of outlets in Africa and the Middle East. Read the Article […]
Kalev spoke yesterday at the Modeling Science, Technology & Innovation Conference held at the National Academies building here in DC. Read More.
Kalev's latest story for Forbes explores how the media list that powers Facebook's Trending Topics module appears to have been constructed as the top 1,000 media properties ranked by total online traffic, necessarily creating an enormous bias towards Western news outlets and the perils of monitoring news versus monitoring society. Read the Full Article.