A GROWING number of machines masquerading as humans have quietly joined the millions of users.
A study has shown that so-called social bots are behind up to 15 per cent of Twitter accounts.
Researchers at Indiana University and the University of Southern California created a bot-tracking system that analyses public data and metadata from Twitter accounts based on more than a thousand characteristics – including friends, content and sentiments expressed in tweets, as well as networking and activity patterns.
If you want to find out if account following is run by a real person or a piece of software designed to tweet like a human, you may want to check out the identification website at truthy.indiana.edu/botornot.
You’ll need to give the tool access to Twitter’s application programming interface (API), which is why the user is asked to sign in.
According to the scientists, however, no information from the account provided will be analysed.
The detection works best with English-language accounts, the experts say, and provides less reliable results for accounts in other languages. (dpa)