Dataset December 2023
Corpus of Contemporary Polish (KWJP)
A balanced and representative corpus of written Polish covering texts from 2011-2020, divided into fiction, non-fiction, and journalism.
Resources
Software November 2021
COMBO
A language-independent NLP system for dependency parsing, part-of-speech tagging, lemmatisation, morphological analysis, and more, built on top of PyTorch and AllenNLP.
Dataset January 2018
Polish Dependency Bank 2.0 (PDB 2.0)
An extended Polish dependency treebank of 22,152 trees and 350,001 tokens, also available in Universal Dependencies format (PDB-UD), with COMBO models trained on it.
Software May 2016
TermoPL
A flexible terminology extraction tool for Polish and other languages, ranking term candidates with the C-value method on top of Universal Dependency structures or a customisable grammar.
Dataset May 2014
Polish Summaries Corpus
A corpus of Polish news summaries based on texts extracted from the Rzeczpospolita corpus, released under the CC BY 3.0 licence. Co-funded by the ATLAS project and by the European Union.
Dataset January 2013
Polish Valence Dictionary (Walenty)
An electronic dictionary of subcategorisation frames for Polish verbs and quasi-verbal predicates, distributed in plain text, PDF, and TEI formats.
Dataset January 2012
National Corpus of Polish (NKJP)
A reference corpus of the Polish language with over 1.5 billion words from classic literature, newspapers, specialist periodicals, conversation transcripts, and internet texts, including a manually annotated 1-million-word balanced subcorpus.