LINGUIST List 32.431

Thu Feb 04 2021

Calls: Slavic Subgroup; Comp Ling, Morphology/Online

Editor for this issue: Lauren Perkins <laurenlinguistlist.org>



Date: 04-Feb-2021
From: Roman Yangarber <roman.yangarberhelsinki.fi>
Subject: Shared Task on Slav-NER: Recognition, Normalization, Classification and Cross-lingual linking of Named Entities in Slavic languages
E-mail this message to a friend

Full Title: Shared Task on Slav-NER: Recognition, Normalization, Classification and Cross-lingual linking of Named Entities in Slavic languages
Short Title: Slav-NER-3

Date: 19-Apr-2021 - 20-Apr-2021
Location: Kyiv (Online), Ukraine
Contact Person: Roman Yangarber
Meeting Email: < click here to access email >
Web Site: http://bsnlp.cs.helsinki.fi/shared-task.html

Linguistic Field(s): Computational Linguistics; Morphology

Language Family(ies): Slavic Subgroup

Call Deadline: 08-Mar-2021

Meeting Description:

The 3rd Slav-NER Shared Task on Named Entities in Slavic Languages: Recognition, Normalization, Classification and Cross-Lingual Linking. Co-located with the BSNLP Workshop, EACL 2021

The 3rd Slav-NER Shared Task focuses on Named Entities in Slavic languages.

Due to rich inflection, free word order, derivation, and other phenomena common to the Slavic languages, work on Named Entities poses important challenges. Fostering research & development on the problems of Named Entities — detecting names, lemmatization (normalization), classification, and cross-lingual matching — is crucial for information access and wider use of NLP in Slavic languages.

The 3rd Slav-NER Shared Task covers six languages:
- Bulgarian,
- Czech,
- Polish,
- Russian,
- Slovene,
- Ukrainian.

and five types of named entities:
- persons,
- locations,
- organizations,
- events,
- products.

For information about training and test data, guidelines, and participation, please see the Shared Task Home Page.

IMPORTANT: Participants are NOT required to perform all tasks or for all languages. For example, a monolingual entry, without lemmatization of the names, can participate.

The Shared Task focuses on cross-lingual extraction of named entities — the systems should recognize, classify, and extract all mentions of a name in a document; detecting the position of each name mention is NOT required. Name mentions should be lemmatized, and mentions referring to the same real-world object should be linked across documents and languages. The text collection consists of sets of documents retrieved from the Web, each set about a certain major entity or event. The corpus was collected by crawling the Web and parsing the HTML documents.

For background, see the 1st (2017) and the 2nd edition (2019) of the Slav-NER shared task.

Final Call for Participation:

Teams that wish to participate should register via email to: bsnlpcs.helsinki.fi, with the following information:
- name of team,
- team members,
- contact person,
- contact email

Important Dates:
Shared task announcement: 1 December 2020 ⇒ Training data available (for most languages)
Release of remaining training data (for Slovene and Ukrainian): 15 January 2021
Registration deadline: 1 March 2021
Release of Test data to registered participants: 3 March 2021
Submission of system responses: 5 March 2021
Results announced to participants: 6 March 2021
Camera-ready shared task papers (optional): 8 March 2021

For additional information, please see the Slav-NER Shared Task home page at http://bsnlp.cs.helsinki.fi/shared-task.html




Page Updated: 04-Feb-2021