Fake news detection engine seeks to combat online harms

Fake news websites, commonly known as hoax news websites, are websites that deliberately publish fake news—hoaxes, propaganda, and disinformation purporting to be real news. The sites often use social media like Facebook, Twitter and Instagram to drive web traffic and amplify their effect.

Unlike news satire, fake news websites deliberately seek to be perceived as legitimate and taken at face value, often for financial or political gain. We live in a time when anyone can just hide behind their keyboard and pretend to be a journalist. it’s becoming increasing difficult to know which news is real and which one is not. Now, universities and research institutions are fighting back with technology.

The University of Exeter is a public research university in Exeter, Devon, South West England, United Kingdom. he college is currently working on developing a digital tool designed to detect fake news, cyberbullying and other online harms.

Dubbed “LOLA,” the tool uses sophisticated artificial intelligence to detect emotional undertones in language, such as anger, fear, joy, love, optimism, pessimism and trust. It can analyze 25,000 texts per minute, and has been found to detect harmful behavior such as cyberbullying, hatred and Islamophobia with up to 98% accuracy.

LOLA takes advantage of the latest advances in natural language processing and behavioral theory. Taking its name from the children’s TV series Charlie and Lola, the detection engine has been developed by a team led by Dr David Lopez, from the Initiative for Digital Economy Exeter (INDEX).

“In the online world the sheer volume of information makes it harder to police and enforce abusive behavior,” said Dr Lopez. “We believe solutions to address online harms will combine human agency with AI-powered technologies that would greatly expand the ability to monitor and police the digital world.

“Our solution relies on the combination of recent advances in natural language processing to train an engine capable of extracting a set of emotions from human conversations (tweets) and behavioral theory to infer online harms arising from these conversations.”

Such is LOLA’s potential in the battle against misinformation that it has already led to collaborations with the Spanish government and Google. In a recent experiment, LOLA was found to pinpoint those responsible for cyberbullying Greta Thunberg on Twitter.

It has also been used to spot fake news about Covid-19, detecting the fear and anger so often used to pedal misinformation and singling out the accounts responsible. LOLA grades each tweet with a severity score, and sequences them: ‘most likely to cause harm’ to ‘least likely’. Those at the top are the tweets which score highest in toxicity, obscenity and insult.

This kind of analysis could be a valuable tool for cybersecurity services, at a time when social media companies are under increasing pressure to tackle online harms. The government is in the process of creating a new regulatory framework for online safety, giving digital platforms a duty of care for their users.

Dr Lopez added: “The ability to compute negative emotions (toxicity, insult, obscenity, threat, identity hatred) in near real time at scale enables digital companies to profile online harm and act pre-emptively before it spreads and causes further damage.”

Dr Lopez gives further details of the project in the document Mitigating Online Harms at Speed and Scale, and more information on LOLA can be found on the INDEX website.