Automated accounts have become more sophisticated and complex in recent years. Many fake accounts are controlled in part by humans, but also by machines, or simply amplify messages written by real people (what Menczer calls “cyborg accounts”). Other accounts use tricks designed to evade human and algorithmic detection, such as quickly liking and disliking tweets or posting and deleting tweets. And of course there are plenty of automated or semi-automated accounts, like those of many companies, that aren’t really malicious.
The Botometer algorithm uses machine learning to assess a wide variety of public data associated with an account — not just the content of tweets, but also when messages are sent, who follows an account, and so on — to determine the likelihood. determine that it is a bot. While the algorithm is state-of-the-art, Menczer says, “many accounts now fall into the range where the algorithm isn’t very sure in principle.”
Menczer and others say that bot spotting is a cat-and-mouse game. But they add that it could become significantly more challenging in the future as spammers use algorithms that are better able to generate persuasive text and have coherent conversations.
Twitter itself is better equipped to spot bots using machine learning, as it has access to a lot more data about each account. This includes a user’s entire activity history, as well as the various IP addresses and devices they use. But Delip Rao, a machine learning expert who worked on spam detection at Twitter from 2011 to 2013, says the company may not be able to reveal how this works, as it could reveal personal data or information that could be used to support its recommendation. manipulate the platform. system.
This week, Musk also got into an argument with Twitter CEO Parag Agrawal over how easily the company could reveal its method of finding bots. On Monday, Agrawal posted a topic explain how complex the challenge still is. He noted that the private data Twitter has could alter the calculations around the number of bots on the service. “FirstnameBunchOfNumbers with no profile picture and strange tweets may seem like a bot or spam to you, but behind the scenes we often see multiple indicators that it is a real person,” he wrote in the thread. Agrawal also said Twitter was unable to release details of these reviews.
If Twitter can’t or won’t reveal its methodology and Musk says he won’t go ahead without details, the deal may remain in limbo. Of course, Musk could use the issue as leverage to negotiate the price.
For now, Musk seems dissatisfied with Twitter’s attempts to explain why finding bots isn’t as easy as he thinks. He responded to Agrawal’s long thread on Monday with: a simple message that seemed much more fitting for a bot than for a potential buyer of Twitter: a single laughing poo emoji.
