Google Chrome extension for the efficient filtration and flagging of webpages on social media containing violent extremist propaganda.
Algorithm uses composite parameters:
- Semantic Analysis - word2vec token clustering
- Tone Analysis - through the IBM Watson API
- Image Filtering - classifying violent imagery and propaganda symbolism
- Source Verification - identification and checking of URLs against a list of verified sources.