WPI researcher targets fake news with $516K federal grant
Kyumin Lee, a professor at WPI, has developed algorithms to target fake news and spam reviews.
Misinformation and fake news spread like wildfire on the internet, influencing public opinion and according to U.S. intelligence agencies, and the 2016 presidential elections.
A researcher at Worcester Polytechnic Institute, however, is using computer science to combat that growing phenomenon, the school said in a press release.
Funded by a $516,000 National Science Foundation award, assistant professor Kyumin Lee has developed algorithms that can detect fake “likes” and followers across different social platforms like Amazon, Facebook and Twitter.
Specifically, Lee’s work targets crowdturfing, which is described as an online black market for false information including fake reviews, damaging tweets and fake news.
Lee, who joined WPI in July, first targeted sites like Amazon’s Mechanical Turk, an online recruiting tool that has been used for crowdturfing campaigns.
Utilizing machine-learning and predictive modeling, Lee’s algorithms sift through posts and searches for patterns his research has shown are associated with illegitimate tasks, including high wages or jobs manipulating or posting information on particular websites or clicking on certain kinds of links.
The algorithm can then identify the organization's, workers and websites posting the material.
Lee’s method can identify new users engaging in crowdturfing, including “liking” content on social media or “following” certain users with 90 percent and 99 percent accuracy.
“The algorithm will potentially prevent future crowdturfing because you can predict what users will be doing,” he said. “Hopefully, companies can apply these algorithms to filter malicious users and malicious content out of their systems in real time. It will make their sites more credible. It’s all about what information we can trust and improving the trustworthiness of cyberspace.”
WBJ Web Partners
Most Recent
Most Recent
0 Comments