WantToKnow.info provides concise summaries of buried/under-reported news stories from reliable media sources. I transformed WantToKnow.info's news summaries database into an easy-to-work-with dataset, and used this as a starting point to begin mapping the connections between news stories. I then applied some basic natural language processing (NLP) including named entity recognition (NER), and created a couple of interactive visualizations using Kumu.io.
1,000 Buried News Stories map - https://kumu.io/Mark-Bailey/1000-buried-news-stories
Ten Years of Economic System Corruption map - https://kumu.io/Mark-Bailey/ten-years-of-economic-system-corruption
Everything was done in python 3.6. You can find the Jupyter notebooks used to build the maps in this video at this repo - https://github.com/ma-da/misc-share/
