When Bots Attack

Someone bought my Twitter account 10K fake/bot followers, and here’s what I learned about Twitter’s spam detection tools

Network graph representation of my Twitter followers (large grey dot is me, Geoff Golberg!). Each dot is one of my Twitter followers. Interactive version is fantastic when viewed on desktop (use touchpad/mouse to move around/zoom, hover over dots to reveal account, click to observe interconnectivity). Graph credit: Max Galka

Story First, Data Second

Attack Of The Bots

MeVee’s Twitter account has changed its handle from @MeVeeApp to @buildwithcrane. The account is now associated with Crane AI, which has nothing to do with live video (MeVee no longer exists)

Well, Shit. This Sucks

Source: TwitterCounter.com (Feb ‘18)

Enough With The Story, Let’s Get To The Data

1) Network Graph Representation

@geoffgolberg’s followers (as of May ‘17)

2) TwitterAudit.com

@geoffgolberg’s followers (“fake” followers as determined by TwitterAudit, Feb ’18)

3) Twitter API

Credit Columbia professor, Mark Hansen, with the fingerprint discovery
@geoffgolberg’s followers (Feb ’18; excludes “suspected spam accounts”)
@geoffgolberg’s followers (Feb ’18; includes “suspected spam accounts”)

Comparing The Three Approaches

Making Sense Of It All

@geoffgolberg’s follower count as of Mar 28th, 2018

OK, So What’s Your Point?

Be sure to read the full thread!


CEO & Founder, Social Forensics | Previously: Co-Founder, Elementus | Featured in BBC, CNN, BuzzFeed, and Quartz, among others | SocialForensics.com