How the tool works

  1. Allow users to search for supporting arguments using an underlying database (news collection) Check a Fact
  2. "Indicate" a Tweet is hate speech based on selected keywords Daily Info
  3. "Indicate" a tweet is fake if it is not covered in our news article collection Daily Info
  4. Provide daily information of such categories (fact, fake, hate) Daily Info
  5. A graph showing the hate speech distribution over several years Yearly Growth
  6. A chart showing the distribution of "hate, fact, and fake" tweets from 2020 onwards. Chart Report

Future plan of the tool

  1. Collect more data to build better "AI" components for the classification
  2. Improve the tool so that it can be used by data journalists to analyze and report their findings
  3. Build a browser plugin that automatically detects disinformation and abusive texts for users