Regarding the possibility the system could be used by fake news creators to assess the potential for false content to pass for real before it is published, Center for Mathematical Sciences Applied to Industry project coordinator Francisco Louzada Neto sai
Credit: Kaspersky.com
A team of researchers at Brazil's University of São Paulo (USP) has developed a Web platform that can automatically detect fake online news.
The platform statistically analyzes writing characteristics of content in Brazilian Portuguese, which is fed into a machine learning-based classifier that can differentiate the linguistic, vocabulary, and semantic patterns of fake and real news.
The classifier can automatically deduce whether submitted content is false; initial tests indicate the system can detect fake news with 96% accuracy.
USP's Francisco Louzada Neto said the project intends "to offer society an additional tool to identify, not only subjectively, whether a news item is false or not."
From ZDNet
View Full Article
Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA
No entries found