‘False news tends to be less grammatical and less factual, with greater reliance on emotionally charged claims, misleading headlines and so on.’
Penn State researchers hope more precise definitions of ‘fake news’ will assist further research, improve detection measures, and lead to corrective actions.
They identify seven categories of ‘fake news’ as accounts that are:
- false
- polarizing
- satire
- misreported
- commentary
- persuasive
- citizen-generated
Classifying fake news can improve training for AI models created to improve detection and also help humans better distinguish real from fake sources.