Researchers at the Universities of Göttingen and Frankfurt and the Jožef Stefan Institute in Ljubljana, have developed a new approach that can help detect fake news leveraging machine learning methods.

As part of the study published in the Journal of the Association for Information Systems, scientists have leveraged machine learning methods to create classification models that can help detect false information.

The model can help identify suspicious messages based on their content and certain linguistic characteristics.

"Here we look at other aspects of the text that makes up the message, such as the comprehensibility of the language and the mood that the text conveys," said Professor Jan Muntermann from the University of Göttingen as quoted in an official release published in Eureaklert!.

The approach is known in principle. For instance, the spam filters that are used by internet companies to identify spam messages.

The key problem with current methods, according to the researchers was that fraudsters are continuously adapting the content and avoiding certain words that are used to identify fake information to avoid being detected.

In the new approach, researchers combined models with high detection rates and robustness. This is so that fake news can still be detected even if certain suspicious words disappear from the text.

"This puts scammers into a dilemma. They can only avoid detection if they change the mood of the text so that it is negative, for instance," explained Dr Michael Siering. "But then they would miss their target of inducing investors to buy certain stocks."

The new approach has a range of applications from market surveillance to criminal prosecutions in the future.

Related Topics