Duo Security has created open source tools and disclosed techniques that can be useful in identifying automated Twitter accounts, which are often used for malicious purposes.
The trusted access solutions provider, which Cisco recently agreed to acquire for $2.35 billion, has collected and studied 88 million Twitter accounts and over half-a-billion tweets. Based on this data, which the company says is one of the largest random datasets of Twitter accounts analyzed to date, researchers were able to create algorithms for differentiating humans from bots.
The dataset, collected using Twitter’s API, includes profile name, tweet and follower count, avatar, bio, content of tweets, and social network connections.
Researchers created their tools and techniques for identifying bots based on 20 unique account characteristics, including the number of digits in a screen name, followers/following ratio, number of tweets and likes relative to the account’s age, number of users mentioned in a tweet, number of tweets with the same content, percentage of tweets with URLs, time between tweets, and average hours tweeted per day.
Tests conducted by experts led to the discovery of a sophisticated cryptocurrency-related scam botnet powered by at least 15,000 bots. These accounts were designed to use deceptive behaviors to avoid automatic detection, while attempting to obtain money from users by spoofing cryptocurrency exchanges, celebrities and news organizations.
Duo Security informed Twitter of its findings. The social media giant says it’s aware of the problem and claims it’s proactively implementing mechanisms to detect problematic accounts.
“Spam and certain forms of automation are against Twitter’s rules. In many cases, spammy content is hidden on Twitter on the basis of automated detections. When spammy content is hidden on Twitter from areas like search and conversations, that may not affect its availability via the API. This means certain types of spam may be visible via Twitter’s API even if it is not visible on Twitter itself. Less than 5% of Twitter accounts are spam-related,” Twitter said.
Duo Security has published a 46-page research paper describing its findings and techniques. The company will release its tools as open source on August 8 at the Black Hat conference in Las Vegas.
“Malicious bot detection and prevention is a cat-and-mouse game,” explained Duo Principal R&D Engineer Jordan Wright. “We anticipate that enlisting the help of the research community will enable discovery of new and improving techniques for tracking bots. However, this is a more complex problem than many realize, and as our paper shows, there is still work to be done.”
Related: Twitter Unveils New Processes for Fighting Spam, Bots
Related: Surge in Anonymous Asia Twitter Accounts Sparks Bot Fears