Date of Award
2026
Degree Type
Thesis
Degree Name
Master of Science in Computer Science
Department
Computer Science and Statistics
First Advisor
Shaun Wallace
Abstract
We present CALLIOPEION and ARIADNE, a novel solution for automatically generating and visualizing Greek poetic scansion from web-based ancient texts. Scansion is the process of marking rhythm and meter in poetry. It is central to classical scholarship, yet current tools remain incomplete, particularly for Ancient Greek. CALLIOPEION introduces a recursive, context-aware algorithm informed by classics experts’ scanning strategies. We apply CALLIOPEION to 48 chapters from Homer’s Iliad and Odyssey and Hesiod’s Theogony. CALLIOPEION outperforms the Classical Language Toolkit (CLTK) and an LLM baseline, improving recall, F1, and total syllable accuracy across the texts. We designed the total syllable accuracy metric with classical experts to address the unique challenges of Ancient Greek texts and poetic scansion. By open-sourcing CALLIOPEION and ARIADNE, we aim to support broader adoption of Human-AI augmented classical text reconstruction and collect data for future improvements. Our work contributes to web-based Information Extraction, Digital Humanities, and long-term cultural preservation through future computational support for ancient language education and text reconstruction.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Alves Pereira, Francisco Xavier, "CALLIOPEION AND THE ARIADNE WEB APP: HUMAN-AI TOOLS FOR ANCIENT TEXT RECONSTRUCTION OF GREEK POETIC SCANSION" (2026). Open Access Master's Theses. Paper 2712.
https://digitalcommons.uri.edu/theses/2712