MPEDS - Automated Coding of Protest Event Data

Machine-Learning Protest Event Data System

Quick Start
Annotation Interface

MPEDS is distributed as a Docker container. To run it on your computer, you first need to download and install Docker. You’ll also need to install Git LFS, which allows you to download the large-file classifiers and vectorizers.

After that, clone or download the Git repository, navigate to the directory in a terminal, and run docker-compose run mpeds

Example code:

git clone
cd mpeds
docker-compose run mpeds

This will build and launch two linked Docker containers. The first contains the geotagging service CLIFF, which is used to extract information on locations mentioned in the articles. The second container is used to run the main MPEDS program.

The CLIFF container will normally take between 10 and 30 minutes to launch. The main MPEDS container will wait for the CLIFF container to be ready before it launches into a command line prompt. Once the container is ready, MPEDS can be run as a Python package.

To run an example script, type python into the command line prompt. This will generate a file mpeds-output.csv which identifies events which contain protest in an example Lexis-Nexis file, then generate data on a set of variables: form, issue, target, social movement organizations, locations, and size.