Room “Sala Seminari” - Abacus Building (U14)
Some Efforts on Understanding and Identifying Text Phenomena Across Languages
Speaker
prof. Alberto Barron Cedeno
Università di Bologna
Abstract
The development of natural language processing has reached unprecedented success and, perhaps for the first time, it is one of the most pervasive among the artificial intelligence disciplines. Still, multiple tasks remain far from solved, which is particularly true for subjective tasks and languages other than English. In this talk, I will discuss some of our efforts on (a) identifying texts written in different language varieties, (b) identifying malicious contents across languages, and (c) understanding how the subjectivity of certain classification problems can be plugged into prediction systems.
Short Bio
Alberto Barrón-Cedeño is an associate professor at Università di Bologna (Italy). Before that, he was a Scientist at the ALT group of Qatar Computing Research Institute (Qatar; winner of the King Salman International Academy for the Arabic Language Award 2023), and an Alain Bensoussan (MSCA) fellow at Universitat Politècnica de Catalunya (Spain). He obtained his PhD on AI from Universitat Politècnica de València (Spain). Alberto is interested in the automatic analysis of diverse qualities of text, such as originality, relevance, and intent; also across languages. He has co-organised five editions of PAN and six editions of CheckThat! at CLEF, a SemEval shared task and was general chairman of the 2022 edition of CLEF, in Bologna. Currently he serves as co-chair of CLEF and as Specialty Chief Editor of the NLP section of Frontiers in Artificial Intelligence. He has published in the top forums of NLP and IR (e.g., IJCAI, ACL, EMNLP, NAACL, SIGIR, ECIR, IP&M, CL, LRE, KNOSYS).
contact person for this Seminar: leonardo.mariani@unimib.it