FANDEMIA

The FANDEMIA project aims to develop an advanced debunking method that integrates philological, rhetorical, and stylistic text analysis with Natural Language Processing (NLP) and Artificial Intelligence (AI) techniques. By bridging humanities and computational sciences, the project seeks to enhance the detection of disinformation in mainstream journalism, with a particular focus on Italian newspaper articles covering immigration. While existing fake news detection methods predominantly rely on content-based approaches and source credibility assessments, FANDEMIA addresses the linguistic dimension of disinformation. The project builds on the premise that disinformation is difficult to detect automatically because it often shares the same high linguistic register as legitimate journalism. Unlike misinformation, which stems from unintentional inaccuracies, disinformation is deliberately crafted to manipulate readers while maintaining the stylistic and rhetorical coherence of credible news. This complexity necessitates a philological approach, leveraging centuries-old techniques in textual reconstruction, authorship attribution, and stylistic analysis. FANDEMIA will create a corpus of newspaper articles (2014-2024)

and employ Discursive News Values Analysis (DNVA) alongside Key Semantic Domain Analysis to isolate rhetorical patterns specific to disinformation. A combination of qualitative and quantitative methodologies will inform the development of an AI-powered tool capable of assessing the authenticity of news articles based on rhetorical and stylistic markers. This interdisciplinary approach offers a novel framework for combating disinformation, fostering critical thinking, and promoting digital literacy. By making its findings and tools publicly accessible, FANDEMIA aims to contribute to academic research, public awareness, and policy development in the fight against fake news.