However, it became obvious that at least some of his tweets were not his own. And identifying whether a particular tweet came from Trump himself or one of his staffers became a mini obsessions for Andrew McGill, a senior associate editor at U.S. politics magazine The Atlantic.
To this end, McGill developed an algorithm that can identify the tweeter using commonalities in the text—and it can do so with surprising accuracy.
As McGill says in his story: “When tested against a mix of 2016 tweets, it correctly flagged the ones sent from an Android 90 per cent of the time. It’s a bit worse at figuring out when a staffer has tweeted, incorrectly attributing iPhone tweets to Trump around 25 per cent of the time, perhaps because staffers sometimes work to imitate his style.”
McGill used machine learning to identify and parse a few common elements in Trumps tweets, and the results were intriguing.
Read Mcgill’s story on the Atlantic website here
Also, you can check out the Twitterbot’s feed at @TrumpOrNot.