Civil unrest event-relevant Twitter classifier training data

Lewis Mitchell · Data Archiving and Networked Services (DANS) · 2018

This dataset contains tweet ids and label to train the protest-relevance classifier used in Tuke et al, "Pachinko Prediction: A Bayesian method for event prediction from social media data", 2018. The data contains two columns: tweet id and a label classifying the tweet associated with that id as protest-relevant or not. Tweets need to be hydrated from the list of ids before use. See the paper for details on training the classifier.

Read the paper · More papers on PaperTik