Polish Named Entity Recognition Tool

49 Last view: 2020-11-26

Polish Named Entity Recognition Tool

Nerf

http://zil.ipipan.waw.pl/Nerf

ID:

404 Nerf is a statistical tool for Named Entity Recognition (NER) based on the Conditional Random Fields (CRF) modelling method. The tool has been constructed as a part of the National Corpus of Polish project. It has been adapted to recognize tree-like structures of NEs (i.e., with recursivelly embeded NEs) using the Joined Label Tagging (JLT) method. The JLT method is a simple method of encoding NE structures as a sequence of labels. With this method various additional informations about NEs of categorical nature – type, subtype, type of derivation – can be encoded on the level of labels and subsequently recognized using the resultant CRF model. The tool can be configured to use various types of observations during the training and recognition process, for example: lexical informations from textual level, or grammatical informations from morphosyntactic level.

You don’t have the permission to edit this resource.

DistributionAvailability

Available - Restricted Use

Licence

GPL

Restrictions: Share Alike

Fee: free of charge

Download location: hidden

Distribution Access/Medium: Downloadable

Contact Person

Jakub Waszczuk

toolService

Tool

Language Independent

Input

Media type: Text

Resource type: Language Description

Modality: Written Language

Output

Media type: Text

Resource type: Lexical Conceptual Resource

Modality: Written Language

Operation

Operating system: Linux

Required Software

Python (version 2.6 or higher)
Cython (version 0.14 or higher)

Evaluation

Evaluated: True

Evaluation level: Diagnostic

Evaluation type: Black Box

Evaluation criteria: Intrinsic

Evaluation measure: Automatic

Evaluation Report: Results (which will be described in the NKJP book; earlier evaluation results can be found in the "Tools and methodologies for annotating syntax and named entities in the National Corpus of Polish", Proceedings of the 2010 International Multiconference on Computer Science and Information Technology (IMCSIT) can be acquired by running cross_validate.py script distributed with the Nerf tool. The process is described in details in the README file, also distributed with Nerf.

Evaluation details: Cross validation of the Nerf tool on the NKJP corpus, with respect to the NKJP Named Entities hierarchy, yielded F-measure of 79%.

Evaluator Jakub Waszczuk

Creation

Programming language: Python, Cython, C

Formalism: Conditional Random Fields, Joined Label Tagging

Resource Creation

Resource Creator

Jakub Waszczuk

Michał Lenart

Creation started: 01/03/2010

Funding Project

National Corpus of Polish (NKJP)

URL: http://nkjp.pl/

Funding Type: National Funds

Funder: Polish Ministry of Science and Higher Education

Funding Country: Poland

Project duration: 01/01/2007 - 30/06/2011

Metadata

Created: 24/11/2011

Last Updated: 24/11/2011

Source: CESAR

Metadata Creator

Jakub Waszczuk

Version

Version: 0.2

Last Updated: 11/10/2011

Usage

Foreseen UseNlp Applications

Use NLP Specific: Named Entity Recognition

Actual Use - Nlp Applications

Use NLP Specific: Parsing

Details: Named entity-based mention detection for the Polish coreference resolution module.

Usage Report

Ogrodniczuk M., Kopeć M. Rule-based coreference resolution module for Polish. In Proceedings of the 8th Discourse Anaphora and Anaphor Resolution Colloquium (DAARC 2011), pp. 191–200. Faro, Portugal.

Used in

Computer-based methods for coreference resolution in Polish texts (CORE)

URL: http://zil.ipipan.wa...

Funding Type: National Funds

Funder: National Science Centre (100%)

Funding Country: Poland

Project duration: 18/04/2011 - 17/04/2014

Actual Use - Nlp Applications

Use NLP Specific: Named Entity Recognition

Details: Recognition of Named Entities for the UIMA Language Processing Chain in ATLAS CMS.

Usage Report

Ogrodniczuk M., Przepiórkowski A. Polish Language Processing Chains for Multilingual Information Systems. G. Bouma et al. (ed.): NLDB 2012, LNCS 7337, pp. 152–157. Springer, Heidelberg.

Used in

Applied Technology for Language-Aided CMS (ATLAS)

URL: http://www.atlasproj...

Funding Types: Eu Funds, National Funds

Funders: European Commission (50%), The Polish Ministry of Science and Higher Education (50%)

Funding Country: Poland

Project duration: 01/03/2010 - 28/02/2013

Actual Use - Nlp Applications

Use NLP Specific: Named Entity Recognition

Details: Recognition of Named Entities in the National Corpus of Polish. Tool trained on the manually annotated million-word subcorpus has been used to annotate the entire NKJP corpus.

Usage Report

Agata Savary, Jakub Waszczuk and Adam Przepiórkowski. 2010. Towards the Annotation of Named Entities in the National Corpus of Polish. In Proceedings of the Seventh International Conference on Language Resources and Evaluation, LREC 2010, Valletta, Malta. ELRA.

Jakub Waszczuk, Katarzyna Głowińska, Agata Savary and Adam Przepiórkowski. 2010. Tools and Methodologies for Annotating Syntax and Named Entities in the National Corpus of Polish. In Proceedings of the International Multiconference on Computer Science and Information Technology (IMCSIT 2010): Computational Linguistics – Applications (CLA’10), pages 531–539, Wisła, Poland. PTI.

Used in

National Corpus of Polish (NKJP)

URL: http://www.nkjp.pl

Funding Type: National Funds

Funder: The Polish Ministry of Science and Higher Education (100%)

Funding Country: Poland

Project duration: 13/12/2007 - 12/06/2011

Documentation

Tool Documentation: Help Functions

Jakub Waszczuk, Katarzyna Głowińska, Agata Savary, Adam Przepiórkowski, "Tools and methodologies for annotating syntax and named entities in the National Corpus of Polish", Proceedings of the 2010 International Multiconference on Computer Science and Information Technology (IMCSIT).

People who looked at this resource also viewed the following:

Resources from the same project

Resources from the same creators