Marrying Universal Dependencies and Universal Morphology

TL;DR

This paper introduces a deterministic mapping from Universal Dependencies to UniMorph, enabling interoperability between the two schemas and enhancing their utility for linguistic tasks, while critically evaluating their respective foundations.

Contribution

The paper provides the first deterministic mapping from UD features to UniMorph schema, facilitating cross-resource validation and combined use for linguistic applications.

Findings

Achieved 64.13% macro-average recall in mapping validation

Identified data scarcity issues affecting compatibility

Provided a critical evaluation of UD and UniMorph schemas

Abstract

The Universal Dependencies (UD) and Universal Morphology (UniMorph) projects each present schemata for annotating the morphosyntactic details of language. Each project also provides corpora of annotated text in many languages - UD at the token level and UniMorph at the type level. As each corpus is built by different annotators, language-specific decisions hinder the goal of universal schemata. With compatibility of tags, each project's annotations could be used to validate the other's. Additionally, the availability of both type- and token-level resources would be a boon to tasks such as parsing and homograph disambiguation. To ease this interoperability, we present a deterministic mapping from Universal Dependencies v2 features into the UniMorph schema. We validate our approach by lookup in the UniMorph corpora and find a macro-average of 64.13% recall. We also note incompatibilities due to paucity of data on either side. Finally, we present a critical evaluation of the foundations, strengths, and weaknesses of the two annotation projects.