Saved in:
Bibliographic Details
Main Authors: Amirzadeh, Hamidreza, Jafari, Sadegh, Harju, Anika, van der Goot, Rob
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.17373
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916411579826176
author Amirzadeh, Hamidreza
Jafari, Sadegh
Harju, Anika
van der Goot, Rob
author_facet Amirzadeh, Hamidreza
Jafari, Sadegh
Harju, Anika
van der Goot, Rob
contents Language typology databases enhance multi-lingual Natural Language Processing (NLP) by improving model adaptability to diverse linguistic structures. The widely-used lang2vec toolkit integrates several such databases, but its coverage remains limited at 28.9\%. Previous work on automatically increasing coverage predicts missing values based on features from other languages or focuses on single features, we propose to use textual data for better-informed feature prediction. To this end, we introduce a multi-lingual Part-of-Speech (POS) tagger, achieving over 70\% accuracy across 1,749 languages, and experiment with external statistical features and a variety of machine learning algorithms. We also introduce a more realistic evaluation setup, focusing on likely to be missing typology features, and show that our approach outperforms previous work in both setups.
format Preprint
id arxiv_https___arxiv_org_abs_2409_17373
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle data2lang2vec: Data Driven Typological Features Completion
Amirzadeh, Hamidreza
Jafari, Sadegh
Harju, Anika
van der Goot, Rob
Computation and Language
Language typology databases enhance multi-lingual Natural Language Processing (NLP) by improving model adaptability to diverse linguistic structures. The widely-used lang2vec toolkit integrates several such databases, but its coverage remains limited at 28.9\%. Previous work on automatically increasing coverage predicts missing values based on features from other languages or focuses on single features, we propose to use textual data for better-informed feature prediction. To this end, we introduce a multi-lingual Part-of-Speech (POS) tagger, achieving over 70\% accuracy across 1,749 languages, and experiment with external statistical features and a variety of machine learning algorithms. We also introduce a more realistic evaluation setup, focusing on likely to be missing typology features, and show that our approach outperforms previous work in both setups.
title data2lang2vec: Data Driven Typological Features Completion
topic Computation and Language
url https://arxiv.org/abs/2409.17373