IIT Home Page CNR Home Page

Seminario: "Detection of Propaganda in the News", Aula 27 - 20 Dicembre 2019 ore 15.00

Many recent political events, like the 2016 U.S.A. or the 2018 Brazilian elections, have raised the attention of institutions and of the general public on the role of Internet and social media in influencing the outcome of these events. Understandably, the aspect of such campaigns that has caught more attention is the spread of fake news. However, faking facts is only one of the many devices that can be used to reach the ultimate goal of persuading and influencing readers’ opinions, i.e. to set a propaganda campaign. 

Internet and social media have allowed cross-border computational propaganda by foreign states or even private organisations to influence the outcome of important public events. 
One of the key components of a propaganda campaign is the use of persuasive messages. In this talk I will analyse those psychological and rhetorical techniques that are the basic building blocks for conveying a persuasive message. Such techniques include the use of logical fallacies and appealing to the emotions of the audience. We will present a novel corpus annotated at the fragment level with such techniques and a deep-learning-based approach to identify them.


From 20/12/2019-15.00 to 20/12/2019-16.00 , Aula 27 - Area della Ricerca CNR di Pisa

Speaker: Giovanni Da San Martino - Qatar Computing Research Institute

Responsible: Stefano Cresci

Responsible: Adriana Lazzaroni


Note: Short Bio Relatore: Giovanni Da San Martino is a Scientist at Qatar Computing Research Institute (https://www.hbku.edu.qa/en/about-qcri), HBKU, since 2014. He received his Master Degree in Computer science from the University of Pisa in 2006 and his Ph.D. in Computer science from the University of Bologna (Italy) in 2009 with a thesis on machine learning techniques for structured data. His research interests are at the intersection of machine learning and natural language processing, with applications to (community) question answering and news analysis. He has co-authored more than 60 papers on those subjects. Since 2018, he is involved in the Tanbih and Catalyst projects, both concerning the building of intelligent tools for the automatic analysis of news; the projects are in collaboration with MIT-CSAIL and media partners such as Al Jazeera and Associated Press, respectively. Inside those projects, his focus is in building automatic systems to detect the use of propaganda in written texts.