Analysis and report on the spread on misinformation on Twitter, using a machine learning algorithm.
Datasets were too large to upload on Github. They can be found here: https://botometer.iuni.iu.edu/bot-repository/datasets.html
Under this name (on that page): cresci-2017
Description: A dataset of (i) genuine, (ii) traditional, and (iii) social spambot Twitter accounts, annotated by CrowdFlower contributors. Released in CSV format.
- AKelly_bot_classify.pdf
Threat assesment and explanation of methods for classifying bot accounts on twitter using the above datasets.
Using k-means clustering to classify objects into groups.
Twitter developers account still under Twitter's approval process!