Twitter recently took steps to rid its site of automated accounts, a.ka. bots, and while that may ease the false information that spreads across social media, their actions were likely far from perfect, and we’re sure there are still plenty of automated accounts on there.
So, how do you know if you’re dealing with a fake account? Well, thanks to a contest run by the Defense Advanced Research Project Agency, some key signs that accounts are fake were recognized, so next time you’re get into a vigorous political fight with an account you don’t know, you may want to keep them in mind.
Signals to watch out for include:
User Profile – The simplest indication that an account is fake is a lack of photo, link or any bio in their profile, although some bots do use stolen photos and more.
Tweet Syntax – If posts seem to be formulaic and/or repetitive there’s a good chance it’s a bot. Plus, if they don’t get an obvious joke, or quickly change a subject, that’s another sign.
Tweet Semantics – If an account seems to overly obsess about a specific topic, like posting a link multiple times, it could be a bot since they are created with a specific purpose in mind.
Temporal Behavior – If the account tweets at an abnormal rate, too regularly, or even at odd times, that can be a sign.
Network Features – Although not a obvious to users, if you notice they only follow a few accounts, or are followed by other bots, it’s likely a bot. Also if their tweets don’t seem to match up with their connections, it may be an indication of no real social interaction.