Report

The whole research work and findings

Data

Our raw, labeled, and engineered data

To access our datasets, kindly use the contact field and indicate which dataset you would like to receive

Project Step 3: Data Collection

  • Dataset 1: 170K Records from Telegram Channel of Sputnik Arabic (AR)
  • Dataset 2: 270K Records from Telegram Channel of Russia Today Arabic (AR)

Project Step 4: Data selection and preparation

  • Dataset 3: 4000 unique records from Sputnik and RT – top views and forwarded – AR
  • Dataset 4: 4000 translated unique records from Sputnik and RT – top views and forwarded – ENG
  • Dataset 5: 500 posts with the most engagements prepared for human labelling (combined dataset Sputnik and RT)

Project Step 6: Human labelling and research findings

  • Dataset 6: 275 posts humanly labelled and connected to narratives

Project step 8: Machine Learning Experiment

  • Dataset 7: Machine labelling of 275 posts from human dataset for comparison and calculated evaluation metrics
  • Dataset 8: Machine labelling of ~1500 additional posts for human evaluation with categorical labels from (partially) correct to (partially) incorrect and explanations of the choice