So how do we spot these accounts in the wild? Following are a number of traits we’ve found in our research. As you might expect, many accounts that are not bots or sockpuppets exhibit some of these traits. None of them are foolproof. But the more of these traits an account displays, the more likely it is to be a disinformation account. In our research, we’ve found it far more helpful to look for evidence of these traits in a large collection of tweets, rather than trying to come up with discrete lists of bots, sockpuppets, trolls, and regular users. It’s often these traits that are most dangerous, and it’s these traits that we can look out for when engaging information online ― and when sharing information ourselves. It is also worth highlighting that many of the traits exhibited by bots and sockpuppets are pulled directly from tactics used in online harassment.
via https://medium.com/data-for-democracy/spot-a-bot-identifying-automation-and-disinformation-on-social-media–2966ad93a203