Márton Kardos


2024

pdf bib
Canonical Status and Literary Influence: A Comparative Study of Danish Novels from the Modern Breakthrough (1870–1900)
Pascale Feldkamp | Alie Lassche | Jan Kostkan | Márton Kardos | Kenneth Enevoldsen | Katrine Baunvig | Kristoffer Nielbo
Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities

We examine the relationship between the canonization of Danish novels and their textual innovation and influence, taking the Danish Modern Breakthrough era (1870–1900) as a case study. We evaluate whether canonical novels introduced a significant textual novelty in their time, and explore their influence on the overall literary trend of the period. By analyzing the positions of canonical versus non-canonical novels in semantic space, we seek to better understand the link between a novel’s canonical status and its literary impact. Additionally, we examine the overall diversification of Modern Breakthrough novels during this significant period of rising literary readership. We find that canonical novels stand out from both the historical novel genre and non-canonical novels of the period. Our findings on diversification within and across groups indicate that the novels now regarded as canonical served as literary trendsetters of their time.

2023

pdf bib
OdyCy – A general-purpose NLP pipeline for Ancient Greek
Jan Kostkan | Márton Kardos | Jacob Palle Bliddal Mortensen | Kristoffer Laigaard Nielbo
Proceedings of the 7th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature

This paper presents a general-purpose NLP pipeline that achieves state-of-the-art performance on the Ancient Greek Perseus UD Treebank for several tasks (POS Tagging, Morphological Analysis and Dependency Parsing), and close to state-of-the-art performance on the Proiel UD Treebank. Our aim is to provide a reproducible, open source language processing pipeline for Ancient Greek, capable of handling input texts of varying quality. We measure the performance of our model against other comparable tools and then evaluate lemmatization errors.