Bot accounts are the bane of Twitter. The automated accounts that are often characterised by the default egg icon can wreak all sorts of havoc and totally turn the tide on topics that are trending. In a new paper, researchers discovered a bot army of 350,000 accounts that all had one thing in common: a love of Star Wars novels.
Juan Echeverria and Shi Zhou, from University College London, wanted to know more about the mysterious world of Twitterbots and they began parsing different approaches to analyse data. Soon, they noticed a pattern of similar quotes coming from a galaxy far, far away.
Discover Magazine explains how the researchers first noticed the pattern:
Drawing from a random sample of one per cent of Twitter users, and filtering for just English-speaking accounts, they noticed something odd in the distribution of geo-tagged tweets. While most of the tweets came from urban areas in North America and Europe — as expected — a puzzling number seemed to come from uninhabited areas such as the middle of the ocean or the desert. Diving deeper, they found that almost all of these unexpected tweets quoted Star Wars novels, and were active for a very short period in June and July of 2013. They also had very few friends and followers and were all listed as coming from a Windows phone.
All the evidence suspiciously pointed to a bot army that was using pre-written source material to make its bots seem like semi-coherent human beings. In this case, sticking to a single identifiable set of material gave the network away.
Most of the tweets also seemed to end with an incomplete sentence. For example one tweeted, “Luke’s answer was to put on an extra burst of speed. There were only ten metres #separating them now. If he could cover t.” That quote comes from Star Wars: Choices of One. The researchers have identified 11 novels being used for material.
The researchers were able to define accounts related to the “Star Wars Botnet” as having the following attributes:
- They only tweet random quotations from the Star Wars novels. Each tweet contains only one quotation, often with incomplete sentences or broken words at the beginning or at the end.
- The only extra text that can be inserted in a tweet are (1) special hashtags that are associated with earning followers, such as #teamfollowback and #followme; and (2) the hash symbol # inserted in front of a randomly chosen word (including stop words, like “the” and “in”) in order to form a hashtag.
- The bots never retweet or mention any other Twitter user.
- Each bot has created <= 11 tweets in its lifetime.
- Each bot has <= 10 followers and <= 31 friends.
- The bots only choose ‘Twitter for Windows Phone’ as the source of their tweets.
- The user ID of the bots are confined to a narrow range between 1.5 × 109 and 1.6 × 109.
A network of this size could be very effective at spamming users, pushing trends, giving the appearance of greater consensus on an unpopular opinion and many other nefarious practices. But this particular network has ominously just sat there, dormant since 2013.
Echeverria and Zhou believe their methods for identifying these accounts can help other academics and security experts to locate larger networks in the future. But the most immediate result of their findings is raising awareness that this exists and it could be deployed at any moment.
So, in the future, if you see a Twitter egg arguing hard for the underrated humanitarian efforts of Putin and it previously just seemed to be a big Star Wars fan, there’s a high likelihood that “it’s a trap!” [Discover Magazine]