Algorithm targets fake reviews and online trolls
Image credit: Colin/Wikimedia Commons
A text analysing algorithm can reveal whether comments or reviews posted online come from the same person using several accounts or whether they are genuine.
Using multiple accounts online to post opinions is frequently used by internet trolls trying to pretend that there is a consensus behind their opinion or by businesses hiring review writers to write multiple reviews to persuade consumers about the strengths of a product.
In business, the method, known as astroturfing, is legal but ethically questionable. The new algorithm by researchers from the University of Texas at San Antonio (UTSA) analyses the details of the writing style in the posts and comments and detects similarities that give away the authorship. The researchers said that even if they try, writers can’t completely conceal their style including word choice and punctuation.
“We hope to develop tools to detect astroturfers so that social media users can make informed choices and resist online social manipulation and propaganda,” said Kim-Kwang Raymond Choo, associate professor of information systems and cyber-security at UTSA.
The term astroturfing comes from a brand of synthetic grass carpeting Astro Turf. It refers to the pretence of the companies trying to persuade the public that the opinions in the solicited articles and reviews are genuinely coming from grassroots participants.
The UTSA team tested their algorithm by analysing posts by some of the most prolific online commentators on various news websites. They found that posts made from multiple accounts and supporting the same opinion would be commonly written in a very similar style that the algorithm would identify as coming from a single writer.
“Astroturfing is legal, but it’s questionable ethically,” said Choo who collaborated on the study with Helen Ashman, associate professor of information technology and mathematical sciences at the University of South Australia, and his two former students Jian Peng and Sam Detchon.
Politicians are frequently being accused of astroturfing. Both current US presidential candidates Donald Trump and Hillary Clinton have been suspected of hiring paid writers to post enthusiastic comments about their proposed policies.
“It can be used for any number of reasons,” Choo said. “Businesses can use this to encourage support for their products or services, or to sabotage other competing companies by spreading negative opinions through false identities.”
Now that Choo has the capability to detect one person pretending to be many online, he is considering further applications for the algorithm such as preventing plagiarism and contract cheating.
Sign up to the E&T News e-mail to get great stories like this delivered direct to your inbox every day.