Twitter bots and how to detect them
In my novice days on Twitter, I was surprised to log in one day to find “the official” Keanu Reeves among my followers. For a nanosecond I felt flattered that a famous movie star should find little old me worth following. Reeves grew up in Toronto, barely two hours from where I grew up, and I often visit Toronto. Even so, it seemed unlikely our paths had ever crossed and more unlikely that I would forget having met him.
Fortunately, I had just learned about #bots — those annoying spam accounts that target and follow legitimate Twitter accounts.
Initially a bot account might appear to be an actual user: along with the word “official”, that Twitter profile photo certainly appeared to be of the real Keanu Reeves, and likely was.
Fortunately there are numerous ways to tell a bot from a legitimate account. These are only some:
1. According to MIT Technology Review https://bit.ly/2zM8KFy, the most common way to tell if an account is fake is to check out the profile. The most rudimentary bots lack a photo, a link, or any bio. More sophisticated ones might use a photo stolen from the web, or an automatically generated account name.
Since that first time, I’ve had to deal with numerous other fake Keanu Reeves accounts, each profile name followed by a different set of computer generated numbers.
According to Chris Makara the absence of a profile image or the use of a stock image or a profile image shared by other fake accounts suggests an account might be fake.
To find out if a photo image has been used by others, check out Makara’s article https://bit.ly/2Ti0Vyu. I’ve used Google Images for this purpose for quite some time.
2. Check the bio and tweets. If they read like they belong to a real person (not delivered by a computer), it’s a good sign it’s not a bot.
3. Look at the history. The Mashable website https://bit.ly/36bFfZM cites this from Robhat labs, the team behind likely-bot-identifier botcheck.me:
"Behavior such as tweeting every few minutes in a full day," the group explains, "endorsing polarizing political propaganda (including fake news), obtaining a large follower account in a relatively small time span, and constant retweeting/promoting other high-confidence bot accounts are all traits that lead to high-confidence bot accounts.”
4. When I’m pretty sure my follower is a bot, I also look at the list of his/her/it’s followers. I’m a woman, and if I find only women listed, I immediately hit BLOCK. In a weak attempt to disguise some bad intentions, the bot will sometimes sprinkle 1 or 2 men usually near the top of the follower list, but don’t be fooled.
5. Bots post a lot, writes Ben Nimmo, Senior Fellow at the Atlantic Council’s Digital Forensic Research Lab as cited on The Internet Citizen website https://mzl.la/2AJM9Kt.
“If you suspect an account might be a bot, the first and easiest thing you can do is check their Twitter activity. Go to the account’s profile page, and see how many tweets they’ve posted since their account was created. A human Twitter user might post 10-15 times a day. A bot account will post with high frequency, up to 2,000 times a day.”
6. Another tip from Ben Nimmo: One of the main roles of bots is to boost up the message of other bots, so if you find an endless stream of retweets instead of original posts, it’s likely a bot.
7. According to Symantec Enterprise https://bit.ly/3dYUecu, many Twitter bots have a relatively recent creation date. My alarm bell rings whenever I see the follower has joined in a current month. It leads me to look for other bot signs. Three signs and I hit BLOCK.
With daily diligence and knowing what to look for, I have become adept at weeding out bots according to results from various bot-detecting applications I’ve used.
Symantec Enterprise lists the following bot-detecting applications that are available for public use:
Botometer: an application developed by the Indiana University Network Science Institute (IUNI) and the Center for Complex Networks and Systems Research (CNetS).
Botcheck.me: a Chrome extension developed by two students at the University of California, Berkley. It specifically detects political propaganda bots.
Tweetbotornot: an open-source package for developers created by Michael Kearney, a professor at the Informatics Institute in the University of Missouri.
Symantec Enterprise (see above URL) also lists several steps Twitter is taking to get rid of fake, spammy, and bot accounts.
Along with blocking a bot account, I now also report the account to Twitter. But alas, the fake Keanu Reeves keeps returning. If only Twitter could stop fake KR at the gate, the Twitter community would be happier.
One final point: The blue verified badge on Twitter lets people know that an account of public interest is authentic. The badge appears next to the name on an account’s profile and next to the account name in search results. It is always the same color and placed in the same location, regardless of profile or theme color customizations. https://bit.ly/2zUgKEn