ju-bezdek / conll2003-sk-ner

Translated version of original conll2003 dataset

annotations_creators	language_creators	languages	licenses	multilinguality	pretty_name	size_categories	source_datasets	task_categories	task_ids
machine-generated expert-generated	found	sk	unknown	monolingual	conll-2003-sk-ner	10K<n<100K	extended|conll2003	structure-prediction	named-entity-recognition part-of-speech-tagging

Dataset Card for [Dataset Name]

Table of Contents

• Dataset Card for [Dataset Name]
  • Table of Contents
  • Dataset Description
    • Supported Tasks and Leaderboards
    • Languages
  • Dataset Structure
    • Data Splits
  • Dataset Creation
    • Source Data
    • Annotations
      • Annotation process

Dataset Description

This is translated version of the original CONLL2003 dataset (translated from English to Slovak via Google translate) Annotation was done mostly automatically with word matching scripts. Records where some tags were not matched, were annotated manually (10%) Unlike the original Conll2003 dataset, this one contains only NER tags

• **Point of Contact: @ju-bezdek **

Supported Tasks and Leaderboards

NER

labels:

• 0: O
• 1: B-PER
• 2: I-PER
• 3: B-ORG
• 4: I-ORG
• 5: B-LOC
• 6: I-LOC
• 7: B-MISC
• 8: I-MISC

Languages

sk

Dataset Structure

Data Splits

train, test, val

Dataset Creation

Source Data

https://huggingface.co/datasets/conll2003

Annotations

Annotation process

• Machine Translation
• Machine pairing tags with reverse translation, and hardcoded rules (including phrase regex matching etc.)
• Manual annotation of records that couldn't be automatically matched