Prague Dependency Treebank - Consolidated 1.0

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 5208–5218 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 5208 Prague Dependency Treebank - Consolidated 1.0 Jan Hajič, Eduard Bejček, Jaroslava Hlaváčová, Marie Mikulová, Milan Straka, Jan Štěpánek, Barbora Štěpánková Charles University, Faculty of Mathematics and Physics, Institute of Formal and Applied Linguistics Malostranské náměstí 25, 118 00 Prague 1, Czech Republic {hajic,hlavacova,mikulova,straka,stepankova}@ufal.mff.cuni.cz Abstract We present a richly annotated and genre-diversified language resource, the Prague Dependency Treebank-Consolidated 1.0 (PDT-C 1.0), the purpose of which is - as it always been the case for the family of the Prague Dependency Treebanks - to serve both as a training data for various types of NLP tasks as well as for linguistically-oriented research. PDT-C 1.0 contains four different datasets of Czech, uniformly annotated using the standard PDT scheme (albeit not everything is annotated manually, as we describe in detail here). The texts come from different sources: daily newspaper articles, Czech translation of the Wall Street Journal, transcribed dialogs and a small amount of user-generated, short, often non-standard language segments typed into a web translator. Altogether, the treebank contains around 180,000 sentences with their morphological, surface and deep syntactic annotation. The diversity of the texts and annotations should serve well the NLP applications as well as it is an invaluable resource for linguistic research, including comparative studies regarding texts of different genres. The corpus is publicly and freely available. Keywords: language resource, textual corpus, treebank, morphology, syntax, semantics, lexicon 1. Introduction In this paper, we present a richly annotated and genre- diversified language resource, the Prague Dependency Treebank-Consolidated version 1.0 (PDT-C in the sequel). PDT-C (Hajič et al., 2020)1 is a treebank from the family of PDT-style corpora developed in Prague (for more infor- mation, see Hajič et al. (2017)). The main features of this annotation style are: • it is based on a well-developed dependency syntax the- ory known as the Functional Generative Description (Sgall et al., 1986), • interlinked hierarchical layers of annotation, • deep syntactic layer with certain semantic features. From 2001, when the first PDT-treebank was published, various branches of PDT-style corpora have been developed with different volumes of manual annotation on varied types of texts, differing in both the original language and genre specification. The treebanks were built with different inten- tion. Manual annotation included in them covered a certain part of the PDT-annotation scheme (see Sect. 3). In the PDT-C project, we integrate four genre-diversified PDT-style corpora of Czech texts (see Sect. 4) into one consolidated edition uniformly and manually annotated at the morphological, surface and deep syntactic layers (see Sect. 5). In the current PDT-C 1.0 release, manual annota- tion has been fully performed at the lowest morphological layer (lemmatization and tagging); also, basic phenomena of the annotation at the highest deep syntactic layer (struc- ture, functions, valency) have been done manually in all four datasets.2 1https://ufal.mff.cuni.cz/pdt-c 2Consolidation and fully manual annotation of the surface- syntactic layer is planned …

Content truncated to 500 words