A New Open-Source Tool for Tracking Disinformation

Mozilla

By Mozilla | Sept. 18, 2020 | Fellowships & Awards

The Social Media Analysis Toolkit (SMAT) — co-created by Mozilla Fellow Emmi Bevensee — provides a free, open, and intuitive way to scrutinize what’s trending on internet platforms


The conversations that trend on internet platforms shape our world in consequential ways, from who we vote for, to what news we read, to how we respond to a pandemic.

But frequently, these conversations don’t trend organically — they’re the result of influence campaigns intended to misinform, radicalize, or polarize. Rather than public opinion influencing what trends, social media trends influence public opinion.

So today, Mozilla Fellow Emmi Bevensee is co-launching the Social Media Analysis Toolkit (SMAT) — a free, intuitive, and open-source tool for scrutinizing online conversations. The tool is available in English and Spanish.

SMAT uses intuitive UX and crisp data visualizations to unpack what’s being said on Twitter, Reddit, 4chan, and 8chan (Insight into other platforms, like Telegram, Parler, and Gab will be available soon.) SMAT’s database structure means that results load instantly; on similar tools, loading can be slow due to integration with various APIs. All results can be exported via CSV. The SMAT team is also developing publicly-accessible tools for monitoring the U.S. elections, such as interactive network graphs of politicians that help to map things like spheres of influence and partisanship.

Users can study everything from what’s trending during a certain period of time, to who is driving the conversation, to what links are being shared the most, and beyond. For example, a researcher looking into COVID conspiracies can determine when the conspiracy began, who is leading the conversation, and which news stories, blog posts, and other links are appearing most frequently in the online conversation.

Says Bevensee: “Our information ecosystem is in a crisis. It’s becoming harder and harder to trust what we read online, and the implications of this are dire — from real-world violence, to public health crises, to skewed elections.”

Bevensee continues: “Researchers, journalists, and activists face major challenges when interrogating online trends. Platforms are tight-fisted with access and data. And the third-party tools that exist are prohibitively expensive, tailored to benefit brands and professional developers, not the public good. They generally focus on just one platform, not a range of them. SMAT leverages open-source investigation to scrutinize what’s being said online, who’s saying it, and why.”

Our information ecosystem is in a crisis. It’s becoming harder and harder to trust what we read online, and the implications of this are dire.

Emmi Bevensee, Mozilla Fellow

As a Mozilla Fellow, Bevensee’s host org is the Anti-Defamation League. Bevensee created SMAT alongside a team of five other researchers and technologists. SMAT was created in collaboration with NCRI, Pushshift, Open Collective, and iDramaLab, which provide advisory, tooling, funding, and other resources.

How it works

Users can leverage one of four SMAT functions to better understand the information ecosystem:

The Timeline functionality takes a search term and visualizes how often it appeared over a period of time and across platforms.

Timeline


The Hashtag functionality takes a search term and visualizes related hashtags.

Hashtag


The Link Counter functionality takes a search term and visualizes what links are appearing alongside it.

Link Counter


The Activity functionality takes a search term and visualizes which authors or subreddits have the most activity.

ACtivity


More than ever, we need a movement to ensure the internet remains a force for good. Mozilla Fellows are web activists, open-source researchers and scientists, engineers, and technology policy experts who work on the front lines of that movement. Fellows develop new thinking on how to address emerging threats and challenges facing a healthy internet. Learn more at https://foundation.mozilla.org/fellowships/.