nert-nlp/streusle

STREUSLE: a corpus with comprehensive lexical semantic annotation (multiword expressions, supersenses)

59
/ 100
Established

This project provides a meticulously annotated English text corpus, taking raw web review text and enriching it with detailed information about multiword expressions (like 'kick the bucket') and semantic supersenses for nouns, verbs, and prepositions. The output is a structured dataset (in CoNLL-U or JSON) that reveals the grammatical status and meaning categories of individual words and phrases. This is designed for computational linguists and natural language processing researchers who need high-quality, fine-grained semantic data for training models or linguistic analysis.

Available on PyPI.

Use this if you need a richly annotated English text dataset to understand the precise grammatical and semantic roles of words and multiword expressions, especially for research in lexical semantics or building advanced NLP systems.

Not ideal if you're looking for a simple, general-purpose text corpus without deep lexical semantic annotations, or if your primary interest is in high-level sentiment analysis rather than detailed linguistic structure.

computational-linguistics lexical-semantics natural-language-processing-research corpus-linguistics multiword-expressions
No Dependents
Maintenance 6 / 25
Adoption 9 / 25
Maturity 25 / 25
Community 19 / 25

How are scores calculated?

Stars

72

Forks

19

Language

Python

License

CC-BY-SA-4.0

Last pushed

Nov 15, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/nert-nlp/streusle"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.