Ossetic-COT: Designing a morphologically annotated corpus and morphological analyzer for Ossetic

2026-07-06

AI Summary

This study addresses the longstanding scarcity of Universal Dependencies (UD)-compliant morphologically annotated corpora and high-accuracy analytical tools for Ossetic. The authors present the first large-scale, manually annotated UD-compatible corpus for Iron Ossetic, comprising 5,454 sentences and 74,032 tokens. Leveraging this resource, they develop a BERT-based neural morphological analyzer trained specifically on this dataset. Experimental results demonstrate that the proposed analyzer achieves a token-level morphological tagging accuracy of 95.60%, marking a significant advancement in computational linguistic research for Ossetic—a critically low-resource language. This work not only establishes a foundational benchmark for future studies but also showcases the effectiveness of transformer-based models in morphologically rich, under-resourced languages when paired with carefully curated annotated data.

Abstract

In this work we present the first morphologically annotated corpus for Iron Ossetic that conforms to the Universal Dependencies schema. The corpus includes 5454 manually annotated sentences from the Iron Ossetic Corpus of Oral Texts, containing 74032 tokens. We use this corpus to train a BERT-based morphological analyzer. The analyzer achieves tag accuracy of 95.60%.

Problem

• Ossetic
• morphological annotation
• Universal Dependencies
• morphological analyzer
• corpus

Innovation

• morphologically annotated corpus
• Universal Dependencies
• BERT-based morphological analyzer
• Iron Ossetic
• tag accuracy