tustultuses
Graphing the influence of real journalism
Phase 1: Real or Fake Journalists
- Curate a list of 100 real journalists, who at least try to find and tell the truth using professional standards of good sourcing, etc.
- Build an API to query against their name (and slight variations) for real/unknown
Phase 2: Real or Fake Headlines
- Build a tool to crawl top news headlines (on Twitter? Google News?) and mark them real/unknown
- Glue headlines together about the same subject (using Google News API?)
- Search for bylines in all, score story based on presence and number of of solid reporters
- Extend API to answer queries against headlines
Phase 3: Real or Fake Rumors
- Extend crawl to blogs, Twitter, and other news distribution channels
- Score distributors of news based on their real/fake ratio of shared headlines
- Extend API to answer queries on these sources
Phase 4: Realish or Fakish
- Rework core ranking of journalists to a scale, using expert panel scoring them on professional reputation
- Build feedback loop to re-score news creators based on their success at providing truth
- Hard to do, but first thought is to see who else reported the same news in our list and let that news cohort drag them up/down
- Extend list of journalists/news creators