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/
▶️ DTube
▶️ IPFS
Great work on this. I recently did a video somewhat on a similar topic (fake news). MSM has a problem.
Cool - I'll check it out!
Wow. This is something! I'd like to talk with you more about this, @mada. Can you email me at [email protected]? I may like to feature this on a video on my channel.
Thanks, @fedoraonmyhead! I will shoot you an email in the next couple of days.
Greatly appreciating all your work on this project.
Hey thanks: )
Always go to @mada when you want the real news!
If dtube fails to load in a timely manner, you can also find this video here:
Congratulations @mada! You have received a personal award!
Click on the badge to view your Board of Honor.