This article presents a set of standardised corpora of poetry comprising over 330,000 poems in ten languages (Czech, English, French, German, Hungarian, Italian, Portuguese, Russian, Slovenian, and Spanish). Each corpus has been deduplicated, enriched with Universal Dependencies, provided with additional metadata, and converted into a unified json structure.
The study investigates the use of toponyms in Czech lyric and narrative poetry, focusing on the second half of the 19th century and authors associated with the Máj, Ruch and Lumír literary schools. While literary geography and onomastics have gained scholarly attention, the specific role of toponyms in distinguishing between literary genres remains under-explored. Using quantitative methods applied to the PoeTree.cs corpus, the research compares the relative frequency of toponyms in lyric and epic collections by authors who wrote in both genres. The results confirm the initial hypothesis that narrative poetry exhibits a higher density of toponyms than lyric poetry. This suggests that alongside the category of time, the category of space, represented by toponyms, plays a significant role in differentiating these genres. The study further explores the nuances of this relationship through a qualitative discussion of individual authors, considering factors such as genre hybridity, the lyricization of the epic, or polymetry. The conclusion emphasizes the importance of spatial analysis in poetry and proposes directions for future research.
25 July 2026 in Digital Scholarship in the Humanities
Rhyme is deceptively intuitive: what is or is not a rhyme is constructed historically, scholars struggle with rhyme classification, and people disagree on whether two words are rhymed or not. This complicates automated rhymed recognition and evaluation, especially in cross-language research contexts. This article investigates how much training data are needed for reliable unsupervised rhyme recognition using RhymeTagger, a language-independent tool that identifies rhymes based on repeating patterns in poetry corpora. We evaluate its performance across seven languages (Czech, German, English, French, Italian, Russian, and Slovene), examining how training size and language differences affect accuracy. To set a realistic performance benchmark, we assess inter-annotator agreement on a manually annotated subset of poems and analyze factors contributing to disagreement in expert annotations: phonetic similarity between rhyming words and their distance from each other in a poem. We also compare RhymeTagger to three large language models (LLMs) using a one-shot learning strategy. Our findings show that, once provided with sufficient training data, RhymeTagger consistently outperforms human agreement, while LLMs lacking phonetic representation significantly struggle with the task.