Researchers at the Fraunhofer-Gesellschaft research organization in Germany have developed a system that automatically analyzes social media posts and filters out fake news and disinformation.
The tool uses machine learning to analyze content and metadata, and draws on user interaction to optimize the results in real time.
The researchers built libraries comprised of serious news pieces as well as texts that users identified as fake news; this forms the dataset used to train the system.
Metadata helps differentiate between authentic sources of information and fake news, allowing the researchers to build heat maps and graphs of send data, send frequency, and follower networks.
Said Fraunhofer’s Ulrich Schade, “Our software can be personalized and trained to suit the needs of any customer. For public bodies, it can be a useful early warning system.”
View Full Article
Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA
No entries found