Twitter’s recent growth has been a cause for celebration across the tech industry, and the finance industry is pretty happy too. However, there’s still a major problem with bots online — and on Twitter especially. Scientists from the University of Indiana developed a tool to identify these bots, but what’s more interesting is why they felt the need to tackle the bot problem.
The methodology for developing the BotOrNot tool was fairly simple. Drawing from a pool of known bots that were identified in 2011, Emilio Ferrara and others compiled a pool of 2.6 million bot tweets and 3 million human tweets. The researchers analyzed both data sets using over 1,000 data points within an algorithm, including things like retweet rate and @mentions received.
Identifying the differences was not particularly difficult: Humans tend to interact with other accounts more often, while bots have longer usernames and younger accounts. Spam accounts are mostly a nuisance for any user, but social bots could present a clear danger.
According to the study [PDF]:
These bots mislead, exploit and manipulate social media discourse with rumours, spam, malware, misinformation, political astroturf, slander or even just noise. This results in several levels of societal harm. For example, bots used in political astroturf artificially inflate support for a candidate; their activity can endanger democracy by influencing the outcome of elections.
The paper goes on to provide examples where this has happened. Another problem is that social bots can be used to inflate Twitter follower numbers, which can give someone in the public eye unwarranted clout. Other dangerous outcomes include damage to the stock market and spurious rumors.
Ultimately, social bots could be damaging because they are becoming more sophisticated and harder to detect. Bots are also easily used for malicious purposes, and as the paper speculates, any organization with sufficient resources and motivations could dominate a topic or conversation. Bots aren’t just bad for business, they’re bad for society.