MultiCoNER II: Multilingual Complex Named Entity Recognition
Complex and ambiguous named entity recognition across 12 languages. Annotators identify fine-grained entity types including creative works, groups, medical terms, and complex entities that require world knowledge. Based on the SemEval-2023 Task 2 shared task for multilingual complex NER.
Configuration Fileconfig.yaml
This Potato config reproduces the annotation task. Save it as config.yaml and run potato start config.yaml to try it.
# MultiCoNER II: Multilingual Complex Named Entity Recognition
# Based on SemEval-2023 Task 2 (Fetahu et al., SemEval@ACL 2023)
# Paper: https://aclanthology.org/2023.semeval-1.310/
# Dataset: https://multiconer.github.io/dataset
#
# Task: Fine-grained NER across 12 languages with complex entities
# Annotators identify named entities from a fine-grained taxonomy
# that includes creative works, groups, medical terms, and more.
#
# Key Challenges:
# - Complex entities requiring world knowledge (e.g., song titles)
# - Ambiguous entities that could belong to multiple types
# - Multilingual text with code-switching
# - Low-context short sentences from search queries and social media
#
# Entity Types (6 coarse groups; each spans fine-grained subtypes):
# - Person: Scientist, Artist, Athlete, Politician, Cleric, SportsManager, OtherPER
# - Location: Facility, HumanSettlement, Station, OtherLOC
# - Group: MusicalGRP, PublicCORP, PrivateCORP, AerospaceManufacturer,
# SportsGRP, CarManufacturer, ORG (companies/institutions)
# - CreativeWork: VisualWork, MusicalWork, WrittenWork, ArtWork, Software
# - Product: Clothing, Vehicle, Food, Drink, OtherPROD
# - Medical: Medication/Vaccine, MedicalProcedure, AnatomicalStructure, Symptom, Disease
annotation_task_name: "MultiCoNER II: Complex Named Entity Recognition"
task_dir: "."
data_files:
- sample-data.json
item_properties:
id_key: "id"
text_key: "text"
output_annotation_dir: "annotation_output/"
output_annotation_format: "json"
annotation_schemes:
- annotation_type: span
name: entities
description: "Highlight all named entities and assign their fine-grained type"
labels:
- "Person"
- "Location"
- "Group"
- "CreativeWork"
- "Product"
- "Medical"
label_colors:
"Person": "#3b82f6"
"Location": "#22c55e"
"Group": "#f59e0b"
"CreativeWork": "#ec4899"
"Product": "#06b6d4"
"Medical": "#ef4444"
tooltips:
"Person": "People: scientists, artists, athletes, politicians, clerics, sports managers, others (e.g., Albert Einstein, Lamine Yamal)"
"Location": "Facilities, human settlements, stations, other locations (e.g., Tokyo, Elbphilharmonie, Ohio State Reformatory)"
"Group": "Musical groups, public/private corporations, sports groups, aerospace/car manufacturers, and other organizations (e.g., Daft Punk, FC Barcelona, Sony, WHO)"
"CreativeWork": "Visual works, musical works, written works, artworks, software (e.g., The Shawshank Redemption, PlayStation 5 Pro games)"
"Product": "Clothing, vehicles, food, drink, other products (e.g., iPhone, Boeing 747, Snapdragon 8 Gen 3)"
"Medical": "Medications/vaccines, medical procedures, anatomical structures, symptoms, diseases (e.g., COVID-19, remdesivir, rVSV-ZEBOV)"
allow_overlapping: false
html_layout: |
<div style="margin-bottom: 10px; padding: 8px; background: #f0f4ff; border-radius: 6px;">
<strong>Language:</strong> {{language}}
</div>
<div style="padding: 10px; border: 1px solid #ddd; border-radius: 6px; line-height: 1.8; font-size: 16px;">
{{text}}
</div>
allow_all_users: true
instances_per_annotator: 100
annotation_per_instance: 2
allow_skip: true
skip_reason_required: false
Sample Datasample-data.json
[
{
"id": "mconer_001",
"text": "The Shawshank Redemption, directed by Frank Darabont, was filmed at the Ohio State Reformatory in Mansfield.",
"language": "English"
},
{
"id": "mconer_002",
"text": "Le groupe Daft Punk a annonce sa separation apres 28 ans de carriere, laissant des fans du monde entier en deuil.",
"language": "French"
}
]
// ... and 8 more itemsGet This Design
Clone or download from the repository
Quick start:
git clone https://github.com/davidjurgens/potato-showcase.git cd potato-showcase/text/domain-specific/multiconerii-complex-ner potato start config.yaml
Dataset & paper
Fetahu et al., SemEval@ACL 2023
Citation (BibTeX)
@inproceedings{fetahu-etal-2023-semeval,
title = "{S}em{E}val-2023 Task 2: Fine-Grained Multilingual Named Entity Recognition ({M}ulti{C}o{NER} 2)",
author = "Fetahu, Besnik and Kar, Sudipta and Chen, Zhiyu and Rokhlenko, Oleg and Malmasi, Shervin",
booktitle = "Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.semeval-1.310",
pages = "2247--2265"
}Details
Annotation Types
Domain
Use Cases
Tags
Found an issue or want to improve this design?
Open an IssueRelated Designs
Multilingual Characterization and Extraction of Narratives
Multilingual narrative analysis over online news, adapted from SemEval-2025 Task 10. Covers the task's three subtasks: (1) entity framing, labeling the role each named entity plays (protagonist, antagonist, innocent); (2) narrative classification, assigning topic-specific narrative labels to the article; and (3) narrative extraction, writing a short justification for the article's dominant narrative. Original data spans the Ukraine-Russia War and Climate Change domains in Bulgarian, English, Hindi, Portuguese, and Russian.
Multilingual Coreference Resolution (CorefUD)
Multilingual coreference resolution across 17 languages using Universal Dependencies-style annotations. Annotators identify entity mentions (names, nominals, pronouns) and link them into coreference chains. Based on the CorefUD dataset and the CRAC 2023 Shared Task on Multilingual Coreference Resolution.
MultiTACRED Multilingual Relation Extraction Dataset
MultiTACRED machine-translates the TACRED relation extraction dataset into 12 languages, keeping 41 TAC relation types plus no_relation. This Potato config reproduces the entity-pair relation task.