You are here : COLLEGIUM-LYON > Version anglaise > Scientific activities > Talks and lectures
-
Partager cette page
René Venegas - University of Zurich Seminar - ‘Artificial intelligence and linguistic analysis in the detection of textual disinformation in Spanish’
On December 8, 2025
Schönberggasse 11, 8001 Zürich
Contact : marialuisa.gagoiglesias-at-uzh-dot-ch
René Venegas, researcher in residence at the Collegium, will present his research entitled ‘Artificial intelligence and linguistic analysis in the detection of textual disinformation in Spanish’ at the Romance languages and literature seminar held by the Faculty of Philosophy at the University of Zurich.
Disinformation poses a significant threat to public trust and democratic discourse. In the Spanish-speaking world, however, linguistic research on this topic remains limited. Project PLU230017 analyses the textual characteristics of Spanish-language disinformation using the PACTE tool, as well as natural language processing (NLP) and machine learning techniques. Studies have shown that false information tends to use an emotional lexicon, short sentences, fragmented syntax and a simple grammatical structure. Hybrid models based on linguistic features, including a linear SVM with 86% accuracy, even outperform some deep learning models. This work has led to the development of VIGILIA, a platform designed to analyse and visualise disinformation. The project contributes to the development of explainable AI adapted to Spanish, offering useful tools for media literacy and combatting information disorder.
René Venegas is a professor and researcher at Pontificia Universidad Católica de Valparaíso, specializing in academic discourse analysis and teaching, using computational and corpus linguistics..
