Sara Tonelli

Since 2013 I am the head of the Digital Humanities research group at FBK. I am currently involved in the HYBRIDS MSCA network, whose goal is to fight disinformation using AI and human-in-the-loop approaches, as well as in the AI4Trust and AI-CODE H2020 projects, both aimed at supporting fact-checkers with AI technologies for text and image analysis. Between  January 2021 and December 2022 I was also the scientific coordinator of the KID ACTIONS European project, aimed at addressing cyberbullying among children and adolescents through interactive education and gamification.

I have a Phd in Language Sciences from Università Ca' Foscari, Venice. In 2020 I got the national habilitation as Associate Professor (seconda fascia) for the area `Information Systems' 09/H1. I am also a member of ELLIS, the European Laboratory for Learning and Intelligent Systems, and an appointed Honorary Fellow (cultore della materia) in Computational Linguistics L-LIN/01 at Università di Pavia, Italy. I am part of the board of the Italian Association for Computational Linguistics (AILC) and I am also a member of the External Advisory Board of the LT-Bridge project and of the TRIFECTA project. Between 2021 and 2024 I served as Liaison Representative of the ACL Special Interest Group on Language Technologies for the Socio-Economic Sciences and Humanities (SIGHUM).

I was involved in the past in several European projects: ODEUROPA (extraction of olfactory information from historical multilingual texts),  PROTECTOR (hate speech detection in the religious domain), STAND BY ME (misogyny detection and digital education to fight violence against women and girls), SCAN2 (text simplification of small claim procedures), Pescado (FP7 - keyword extraction), Terence (FP7 - event-based text simplification), NewsReader (FP7 - event extraction and semantic role recognition), SIMPATICO (H2020 - text simplification in the administrative domain), HATEMETER (REC - social media monitoring for islamopohobia detection), CREEP (EitDIGITAL - Cyberbullying detection).

In 2025 I will be Program Co-Chair of the 21st Conference on Information and Research Science Connecting to Digital and Library Science (IRDCL2025).  I am Senior Area Chair for the “Offensive and non-inclusive language detection and analysis” track at COLING 2025 and I serve as Internal Communication Chair at ACL 2025. I am also Advisory Board Member of the Computational Humanities Research journal, Oxford University Press.

Students' supervision:

Current Phd Students:

2023 - present: Katarina Laken (co-advised with Marcos García González), "Analysis of the role of disinformation and hate speech in multilingual social environments".  HYBRIDS Marie Skłodowska-Curie Actions.

2023 - present: Siddharth Bhargava (co-advised with Patricia Martín-Rodilla) "Identifying the stance of argumentative opinions in political discourse", HYBRIDS Marie Skłodowska-Curie Actions.

2022 - present: Nicolò Penzo (co-advised with Bruno Lepri and Marco Guerini) "Analysing the effect of counter-narratives on hateful conversations online", ICT Doctoral School, University of Trento

2021 - present: Teresa Paccosi, "Extraction of Olfactory Information from Texts", Doctoral School in Cognitive Science, University of Trento

 

Past Phd students:

2020 - 2024: Camilla Casula "Rise and Pitfalls of Synthetic Data for Abusive Language Detection", ICT Doctoral School, University of Trento

2019 - 2023: Daniela Trotta (co-advised with Annibale Elia) "Multimodal political communication", Dept. of Linguistics, University of Salerno

2018 - 2022: Federico Bonetti, "Gamification for Linguistic Annotation", Doctoral School in Cognitive Science, University of Trento

2017 - 2022: Matteo Lorenzini (co-advised with Marco Rospocher) "Automatic quality improvement and content enrichment of digital cultural heritage data", ICT Doctoral School, University of Trento

2016 - 2020: Lorenzo Lucchini (co-advised with Bruno Lepri), "Modeling and forecasting cultural dynamics with natural language processed data",  ICT Doctoral School, University of Trento

2013 - 2018: Stefano Menini, "Automatic Analysis of Agreement and Disagreement in the Political Domain", ICT Doctoral School, University of Trento

2013 - 2018: Rachele Sprugnoli, "Event detection and classification for the Digital Humanities", ICT Doctoral School, University of Trento

2012 - 2016: Paramita Mirza, "Extracting temporal and causal relations between events", ICT Doctoral School, University of Trento [arXiv]

Social media analysis and Hate speech detection: CREEP and Hatemeter projects, KID ACTIONS project, PROTECTOR project, PERCEPTIONS project.

Event and temporal/causal processing: Terence and NewsReader European projects, Rachele Sprugnoli's thesis, Paramita Mirza's thesis [link to software, poster, download CausalTimeBank]

Agreement, Disagreement and Argumentation Mining, especially in the political domain: collaboration with INRIA & University of Nice, Stefano Menini's thesis [link to data, link to datasets], collaboration with University of Mannheim (honorable mention at EMNLP 2017 for joint paper)

Digital Humanities, Historical Data Processing, Digital Cultural Heritage: ALCIDE project [link to demo], project on Epistolario De Gasperi and Edizione Nazionale Aldo Moro, Matteo Lorenzini's thesis work. Current ODEUROPA project.

Text Simplification: Terence and SIMPATICO European projects, MUSST syntactic simplifier [link to software], ERNESTA simplification tool, SIMPITIKI corpus for Italian Simplification [link to dataset]

FrameNet, Keyword Extraction and Terminology: my Phd thesis work [link to Italian FrameNet data], collaboration with Luciano Serafini to create a FrameNet-based resource for the Semantic Web [link to data], Keyphrase Digger extraction tool [link to demo and software].