How Algorithms Can Help Defend Us Against Social Network Disinformation Campaigns

I’ve long believed that the best way to counter most social media disinformation campaigns is by developing algorithms to detect them. Fake accounts or those designed to stoke viral responses on social networks, along with astroturfing — an attempt to simulate mass support for a slogan and pass it off as a genuine grassroots protest — follow certain patterns and should be easily identified by algorithms.

Now, MIT has developed an algorithm-based tool called Reconnaissance of Influence Operations (RIO) that proves these kinds of disinformation campaigns can be identified by algorithms, automating a task that should help crack down on attempts to sow division and polarization around the world. The conceptual framework integrates natural language processing, machine learning, graph analysis and a networked causal inference approach that can quantify the impact of individuals and organizations in spreading false narratives.

By now, we already have more than enough data to train an algorithm: there are any number of examples of disinformation campaigns around the world that can be traced back to phenomena of this type with specific, recognizable patterns. Differentiating a genuine viral campaign, something that “sets the networks on fire”, from a premeditated campaign designed with the support of specific account networks may seem difficult, especially considering that in many cases, many real accounts help spread disinformation whose message they share. It might seem difficult, but once it is possible to access to the relationship architecture and account information, algorithms are more than up to the task of identifying dissemination patterns.

Identifying these types of campaigns quickly is essential so they can be dealt with the right way, especially at times when the increased polarization they seek can result in reactions beyond social networks, particularly during elections or a periods of social unrest.

Basically, we are talking about controlling an environment that developed naturally, and then became subject to relatively spontaneous manipulation, and that now uses all kinds of sophisticated techniques and mechanisms in the face of the passivity or incompetence of the managers of social networks. This is trickery in a context that is still not easy for us to identify. The sooner we understand how these types of non-genuine phenomena work, the better for everyone.

—

This post was previously published on Medium.

***

You Might Also Like These From The Good Men Project

Compliments Men Want to Hear More Often

Relationships Aren’t Easy, But They’re Worth It

The One Thing Men Want More Than Sex

..A Man’s Kiss Tells You Everything