The last year has seen a lot of talk about interference in western democracies by spreaders of fake news.  Central to these discussions has been the creation of bot armies that can help to spread the fake news pieces created.  As such, there has been a significant focus on how social networks can better identify fake accounts.

A team from New York University propose their own solution in a recently published paper.  They believe they’ve identified key characteristics of Russian Twitter bots.

“There is a great deal of interest in understanding how regimes and political actors use bots in order to influence politics,” the authors say.  “Russia has been at the forefront of trying to shape the online conversation using tools like bots and trolls, so a first step to understanding what Russian bots are doing is to be able to identify them.”

What bots look like

Whilst many bots are evident from the typical profile photo of an attractive young lady, their style of posting also makes their identity clear.  Indeed, even the platform they use to post is a give away.

“Bots are much more likely to use online platforms while humans frequently use mobile devices,” the authors explain. “However, humans and bots are not dramatically different from each other on a number of other features that characterize their tweeting activity—similarities that reveal a relatively high level of bots’ sophistication.”

The researchers analyzed Twitter during two key periods – February 6th 2014 to October 1st 2014, and January 30th 2015 and December 31st 2015.  During this time, we saw the Russian annexation of Crimea, conflict in Ukraine and the murder of Boris Nemtsov, the leader of the Russian opposition.

In total, over 15 million tweets from approximately 230,000 Russian Twitter accounts were analyzed.  The results revealed that nearly 63,000 of those accounts were bots, with those who actively tweeted about politics nearing 50%.  This number rose significantly during the annexation of Crimea.

The researchers also found a number of other traits that tend to mark out automated posts from their natural counterparts:

• Humans are much more likely to use geo-location in their tweets.
• Bots are more likely to retweet something than human users.
• The most common form of bot activity involves them tweeting news headlines without the link to the original source website.

“This suggests that an important strategy in the use of bots for the purposes of propaganda might be to promote specific news stories and news media in the rankings of search engines,” the authors say.

Suffice to say, bots aren’t the sole preserve of the Russians, and numerous governments utilize such technology to manipulate the media, but it is imperative that social networks get better at identifying, and then removing such accounts.  Hopefully this research will help in doing just that.