Misael MONGIOVÌ
Misael Mongiovì is an Associate Professor of Computer Science (SSD INF/01) at the Department of Mathematics and Computer Science of the University of Catania, where he has served since 2023. He also holds a collaboration appointment at the Institute of Cognitive Sciences and Technologies (ISTC) of the Italian National Research Council (CNR). He has obtained the Italian National Scientific Qualification as Full Professor in Computer Science.
He earned his degree in Computer Science with honours from the University of Catania in 2003 and his PhD in Computer Science there in 2007. His career combines industrial experience with an international academic path: he carried out research at Tel Aviv University, the University of California at Santa Barbara, New York University and the University of the Basque Country, before joining the CNR-ISTC as a researcher, where he coordinated the Catania support unit.
His research addresses data mining and management of textual and graph-structured data, with applications to natural language processing, the integration of Large Language Models and knowledge graphs, and the analysis of complex networks. He is an Associate Editor of the Journal of Big Data (Springer) and, among other honours, received a Best Paper Honorable Mention at DATA 2024 and the Best Paper Award at CSB 2009 (Stanford University). At the Department of Humanities he teaches Introduction to Computational Linguistics in the Master's programme in Text Sciences for Digital Professions (LM-43).
Misael Mongiovì graduated in Computer Science with honours (110/110 cum laude) from the University of Catania in 2003 and earned his PhD in Computer Science from the same university in 2007, with a thesis on distributed and probabilistic data mining algorithms applied to bioinformatics and sensor networks. He holds the Italian National Scientific Qualification as Associate Professor (2018) and as Full Professor (2023) in the 01/B1 – Computer Science sector.
Internationally, he conducted research as a postdoctoral fellow at the Blavatnik School of Computer Science, Tel Aviv University (2009–2010) and as a Research Scientist and Postdoctoral Scholar at the Department of Computer Science of the University of California at Santa Barbara (2010–2012), with additional research visits to New York University and the University of the Basque Country. He also held several research fellowships at University of Catania between 2008 and 2015, and in 2020 he took up a permanent researcher position at the CNR-ISTC, coordinating its Catania Support Unit until 2023.
Since 1 October 2023 he has been an Associate Professor of Computer Science at the Department of Mathematics and Computer Science of the University of Catania, while retaining his collaboration appointment with the CNR-ISTC. He is principal investigator of the university-funded PIACERI project on enhancing language-processing models under data scarcity and is a member of the faculty board of the PhD programme in Computer Science. He also holds institutional roles within the Department of Mathematics and Computer Science, including serving as secretary of the Joint Teacher–Student Committee.
His teaching activity at the University of Catania includes the courses of Programming II and Natural Language Processing in the Computer Science degree programme, and Introduction to Computational Linguistics in the Master's programme in Text Sciences for Digital Professions (LM-43) at the Department of Humanities. He teaches PhD courses on Large Language Models and Agentic AI and supervises doctoral students and numerous theses on NLP and artificial intelligence.
He is the author of more than 50 international scientific publications and serves as Associate Editor of the Journal of Big Data (Springer). He has held organising and program-committee roles at major international conferences, including The Web Conference, ECAI and ACL, and received a Best Paper Honorable Mention at DATA 2024 and the Best Paper Award at CSB 2009 at Stanford University.
Misael Mongiovì's research currently focuses on natural language processing and Large Language Models, with particular emphasis on methods that operate effectively under data scarcity. He is the principal investigator of the university-funded PIACERI project, dedicated to enhancing language-processing models when training data are absent or limited.
Within this area he develops generative approaches to hierarchical text classification in low-data regimes (work recognised with a Best Paper Honorable Mention at DATA 2024), grammar-constrained natural language generation, data-augmentation and generative-modelling techniques for entity recognition, and prompting strategies to improve model performance. A significant part of his work concerns the integration of Large Language Models and knowledge graphs: graph-based retrieval over multiple documents, Retrieval-Augmented Generation with entity linking, explainable claim verification through graph alignment, and knowledge-graph-based inference for semi-explainable systems.
A related line investigates the capabilities and limits of language models from a semantic and cognitive standpoint: the representation of moral and emotional content, the use of LLMs for sign-language translation in low-resource contexts, and a critical analysis of inter-annotator disagreement. On these topics he has published in leading international venues (including Findings of ACL, ECAI and Q1 journals) and is an invited speaker on Large Language Models and knowledge representation at an international school in the field.
This research programme builds on solid prior experience in mining and managing textual and graph-structured data — from knowledge extraction and reconciliation from text to the analysis of complex networks — which continues to provide tools and methodological perspectives for his current work on language models.