The viewer is disabled because this dataset repo requires arbitrary Python code execution. Please consider removing the loading script and relying on automated data support (you can use convert_to_parquet from the datasets library). If this is not possible, please open a discussion for direct help.

Dataset Card for [Dataset Name]

Dataset Summary

The Universal Morphology (UniMorph) project is a collaborative effort to improve how NLP handles complex morphology in the world’s languages. The goal of UniMorph is to annotate morphological data in a universal schema that allows an inflected word from any language to be defined by its lexical meaning, typically carried by the lemma, and by a rendering of its inflectional form in terms of a bundle of morphological features from our schema. The specification of the schema is described in Sylak-Glassman (2016).

Supported Tasks and Leaderboards

[More Information Needed]

Languages

The current version of the UniMorph dataset covers 110 languages.

Dataset Structure

Data Instances

Each data instance comprises of a lemma and a set of possible realizations with morphological and meaning annotations. For example:

{'forms': {'Aktionsart': [[], [], [], [], []],
  'Animacy': [[], [], [], [], []],
  ...
  'Finiteness': [[], [], [], [1], []],
  ...
  'Number': [[], [], [0], [], []],
  'Other': [[], [], [], [], []],
  'Part_Of_Speech': [[7], [10], [7], [7], [10]],
  ...
  'Tense': [[1], [1], [0], [], [0]],
  ...
  'word': ['ablated', 'ablated', 'ablates', 'ablate', 'ablating']},
 'lemma': 'ablate'}

Data Fields

Each instance in the dataset has the following fields:

  • lemma: the common lemma for all all_forms
  • forms: all annotated forms for this lemma, with:
    • word: the full word form
    • [category]: a categorical variable denoting one or several tags in a category (several to represent composite tags, originally denoted with A+B). The full list of categories and possible tags for each can be found here

Data Splits

[More Information Needed]

Dataset Creation

Curation Rationale

[More Information Needed]

Source Data

Initial Data Collection and Normalization

[More Information Needed]

Who are the source language producers?

[More Information Needed]

Annotations

Annotation process

[More Information Needed]

Who are the annotators?

[More Information Needed]

Personal and Sensitive Information

[More Information Needed]

Considerations for Using the Data

Social Impact of Dataset

[More Information Needed]

Discussion of Biases

[More Information Needed]

Other Known Limitations

[More Information Needed]

Additional Information

Dataset Curators

[More Information Needed]

Licensing Information

[More Information Needed]

Citation Information

[More Information Needed]

Contributions

Thanks to @yjernite for adding this dataset.

Downloads last month
1,485