AI could be the latest tool against the rise of fake news, as researchers develop a method of detecting clickbait headlines.
A team of researchers have developed a method of automatically spotting clickbait headlines.
The method, pioneered by the International Institute of Information Technology Hyderabad, uses AI to break text down into document vectors, which provide a way of understanding the meaning or similarity of phrases, sentences or paragraphs.
The approach looks at each word in the headline to figure out its significance to the article as a whole, and ultimately gives a score for just how “clickbait-y” the headline is.
The method developed by the researchers was capable of identifying clickbait headlines accurately in more than 83% of cases.
Clickbait, which was added to the Oxford English Dictionary in 2016, refers to attention-grabbing and often slightly misleading online headlines, which are designed to specifically lure visitors to the site.
The International Institute of Information Technology Hyderabad describes clickbait headlines as “purposefully misrepresenting or promising what can be expected while reading a story on the web be it headline, image or related text”.