Media Cloud

ICWSM 2020

Media Cloud at ICWSM 2020

Investigating Attention and Influence Online with Media Cloud

We’ll be hosting a tutorial session at ICWSM 2020 on using Media Cloud to conduct media research and analysis.

Tutorial overview

When: June 8, 2020 at 1:30 PM EDT

Where: A link to participate will be sent out ahead of the session.

What: Today’s media universe encompasses traditional media, digital platforms, social media and myriad tools responsible for story creation and distribution. The complexity of our media ecosystems presents significant challenges for anyone interested in studying information, particularly across platforms and dissemination methods. In this tutorial, we present Media Cloud, an open source research platform that offers easy, unparalleled access to information from the open web. This tutorial will train attendees to use Media Cloud’s suite of tools for their own media research and analysis purposes through a combination of case studies and exercises.


We are planning for a three-hour session that breaks down Media Cloud’s tools with increasing complexity as the session continues.

Section 1: Overview and Introduction

  • Introduction — 15 minutes

  • Break — 5 minutes

  • Introduction to Explorer and Source Manager — 45 minutes

    • Demo on a climate change-related query — 15 minutes

    • Hands-on breakout session — 10 minutes

    • Shareback — 10 minutes

    • Discussion/Q&A — 10 minutes

  • Break — 5 minutes

Section 2: Case Studies on Research Methods

  • Case Studies on Explorer and Topic Mapper — 30 minutes

    • Case studies on methods the Media Cloud team has used for projects on denigrating language related to immigration and mis- and disinformation related to Greta Thunberg

  • Break — 5 minutes

Section 3: Using Topic Mapper for Multi-Platform Projects

  • Multi-platform Research Using Topic Mapper — 30 minutes

    • A walkthrough of Media Cloud’s new workflow for doing multi-platform research

  • Break — 5 minutes

Section 4: Accessing the Media Cloud API

We’ve created a set of Python 3 Jupyter notebooks to take you through Media Cloud’s API while thinking about some key research questions for Media Cloud projects. These are tailored to be used with a hosted Jupyter Lab environment on Bindery during the session, but you can also run the notebooks on your local machine.

  • API Walkthrough — 45 minutes

    • Accessing Explorer data through the API — 10 minutes

    • Hands-on API session — 10 minutes

    • Accessing Topic Mapper data through the API — 10 minutes

    • Hands-on API session — 10 minutes

  • Conclusion and Wrap-Up Questions — 5 minutes



Please do the following prior to the tutorial:

  1. Register for a Media Cloud account here. You’ll use your Media Cloud account to access our tools Explorer, Topic Mapper, and Source Manager, and to obtain an API key.

  2. If you want to run Media Cloud’s API tutorial notebooks on your local machine, you’ll need to install Python 3.x and set up Jupyter notebooks. (If you are unfamiliar with Jupyter notebooks, we suggest going through this tutorial.)

  3. Think about what you’re most interested in researching — brainstorm some ideas for research topics that will be interesting for you to test during this tutorial.


We know that learning online is different than learning in-person. To help make this experience as meaningful as possible, we recommend doing the following:

  • We’ve inserted several sections for breaks and questions throughout the tutorial agenda. If you have further questions as we go, please post them to the chat. We’ll monitor the chat closely to address questions as they arise.

  • Use two monitors if possible, one to try out Media Cloud on your own, and another to follow along with the session as we go. If you don’t have two monitors, we recommend using your primary monitor to work on Media Cloud and following along with the tutorial from a secondary device such as a phone or a tablet.

  • Use headphones for the best audio experience. Headphones can help limit external noise and make it easier to focus.



These are some of the projects that our team and others have been working on lately. We’re excited to discuss the methods used in these projects, and to help you use Media Cloud tools for your own work!

Partisanship, Propaganda, and Disinformation: Online Media and the 2016 U.S. Presidential Election


How The Media Has — And Hasn’t — Covered Tara Reade’s Allegation

Define American:

Major Newspapers Uncritically
Echo Trump Administration Anti-Immigrant Language

A Story in Images: The Abortion Media Storm of 2019


Rahul Bhargava is a researcher and technologist specializing in civic technology and data literacy. He creates interactive websites used by hundreds of thousands, playful educational experiences across the globe, and award- winning visualizations for museum settings. As a Re- search Scientist at the MIT Center for Civic Media, Rahul leads technical development on projects ranging from interfaces for quantitative news analysis, to platforms for crowd-sourced sensing. He serves as Chief Technology Officer at Media Cloud.

Aashka Dave is a researcher/community manager for Media Cloud, based at the MIT Media Lab. Aashka conducts research on media ecosystems, public health and mis- and disinformation with our tools. She also supports the Media Cloud community in their research efforts. Aashka has a background in media research and digital journalism, holding an MS in Comparative Media Studies from MIT. She previously worked at the Harvard Kennedy School and The Associated Press.

Orestis Papakyriakopoulos is researcher at the Technical University of Munich and Visiting Scholar at the Center for Civic Media at the MIT Media Lab. Orestis studies new and old media by the application of data-intensive algorithms. He also studies the political impact of the use of data-intensive algorithms in society.