GDELT's Global Knowledge Graph (GKG) is frequently used as a filter onto the world's media landscape, scanned through its thematic, geographic, person and organization mentions lists to identify relative global coverage that is then analyzed further. One of the challenges to date has been determining whether an article is truly relevant to a specific question. A surge in mentions of flooding damage in a given town might reflect an active natural disaster, an anniversary reflection on a past disaster or merely speculation about the potential impacts of a changing climate on weather-related risks. Similarly, coverage of disease outbreaks can report a breaking epidemic just as easily as it can simply recount historical incidents or theoretic simulations.

As of today the GKG 2.0 now includes the page title for all articles processed after noon EST today in the XMLExtras field in a new <PAGE_TITLE></PAGE_TITLE> block, making it possible to filter based on the title, using statistical or machine learning algorithms to analyze contextual hints in the page's title to determine the relevancy of a given article for a specific analysis. As with all other GKG 2.0 fields, non-ASCII characters are escaped using their corresponding HTML entities.

We're excited to see how you're able to use this new analytic capability!