Title: Machine Learning for Ancient Languages

Speaker: Prof. John Pavlopoulos @ Athens University of Economics and Business

Time : 16:30 ~ 17:30, Oct 2nd, 2024

Location: Online

• https://us02web.zoom.us/j/82710452194?pwd=LhhpAzzHfIUS8to8exYLca59QGZuG0.1 

Language: English speech & English slides

Abstract: 

Ancient languages preserve the cultures and histories of the past. However, their study is fraught with difficulties, and experts must tackle a range of challenging text-based tasks, from deciphering lost languages to restoring damaged inscriptions, to determining the authorship of works of literature. Based on a recent survey and a related workshop, this talk will focus on recent published research using machine learning for the study of ancient texts. By analysing the relevant literature, lessons learnt and promising directions for future work in this interdisciplinary field are highlighted. Selected studies related to this talk are provided below.

Suggested literature: 

[1] Pavlopoulos, John, Maria Konstantinidou, Elpida Perdiki, Isabelle Marthot-Santaniello, Holger Essler, Georgios Vardakas, and Aristidis Likas. "Explainable dating of greek papyri images." Machine Learning (2024): 1-22.

[2] Pavlopoulos, John, Ryan Sandell, Maria Konstantinidou, and Chiara Bozzone. "HoLM: Analyzing the Linguistic Unexpectedness in Homeric Poetry." In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pp. 8166-8172. 2024.

[3] Pavlopoulos, John, and Maria Konstantinidou. "Computational authorship analysis of the Homeric poems." International Journal of Digital Humanities 5, no. 1 (2023): 45-64.

[4] Pavlopoulos, John, Maria Konstantinidou, Isabelle Marthot-Santaniello, Holger Essler, and Asimina Paparigopoulou. "Dating Greek Papyri with text regression." In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 10001-10013. 2023.

[5] Pavlopoulos, John, Alexandros Xenos, and Davide Picca. "Sentiment analysis of Homeric text: The 1st Book of Iliad." In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pp. 7071-7077. 2022. [Disclaimer: A better way to capture polarised annotations is discussed in a recent article about toxic language detection (or watch this presentation).]    

[6] Assael, Yannis, Thea Sommerschield, Brendan Shillingford, Mahyar Bordbar, John Pavlopoulos, Marita Chatzipanagiotou, Ion Androutsopoulos, Jonathan Prag, and Nando de Freitas. "Restoring and attributing ancient texts using deep neural networks." Nature 603, no. 7900 (2022): 280-283.

Bio:

John Pavlopoulos is an Assistant Professor in Machine Learning for Information Retrieval and Data Mining of Large Document Collections at the Department of Informatics, Athens University of Economics and Business, Greece. He also holds affiliations with Archimedes/Athena RC, Greece, and Stockholm University, Sweden. Previously, he served as a Senior Lecturer (fixed-term) in NLP and Data Science at the Department of Computer and Systems Sciences, Stockholm University, Sweden, and was a visiting scholar at the Venice Centre for Digital and Public Humanities, Ca’ Foscari University of Venice, Italy. His research focuses on Machine Learning for Natural Language Processing, with a strong emphasis on applications in Digital Humanities. His work has been published in top-tier journals (e.g., Nature, Machine Learning, Cognitive Computation) and leading NLP conferences (e.g., ACL, EMNLP).