Search

This report focuses on the metadata as a specific type of research data in the humanities by analysing key metadata elements – persistent identifiers (PIDs), abstracts, keywords and citations. It defines those elements, outlines challenges for processing them in the humanities and presents the challenges for GoTriple as the metadata aggregator of this kind of research data. The assumption is that GoTriple is a specific kind of research dataset on its own that can and will be reused by stakeholders such as other metadata aggregators, indexers, publishers, information services (i.e. providers of scholarly metrics), but also scientists interested in data-driven research (cultural analytics, scientometrics, bibliometrics, etc.). This demands a good understanding of key metadata elements important to GoTriple's aggregation and enrichment processes (abstracts, keywords) and their development (PIDs, citations). Chapter 1 defines the aim of the deliverable, context of its creation and its audience. Chapter 2 discusses the specificity of the research data in the humanities and this report’s position in the rich discussions on the topic. Chapter 3 – dedicated to PIDs – presents the overview of the topic and the challenges related to the PID’s uptake by the humanities, such as the role of cultural heritage data for the humanities, importance of bibliodiversity and multilingualism (subchapter 3.1), then it proceeds to the discussion of processing PIDs from GoTriple’s data providers by focusing on data dispersion and heterogeneity (subchapter 3.2). Chapter 4 – dedicated to keywords – begins with the typology of keywords and the expected standards they should adhere to (subchapter 4.1). Subchapter 4.2 tackles the issue of automated generation of keywords and proposes different approaches applicable in the context of GoTriple. In the subchapter 4.3 a current approach to keyword organisation in GoTriple is presented, with focus on the GoTriple vocabulary that responds to the need for keywords LOD-ification and can be in the future reused for automated keyword generation. Chapter 5 – dedicated to abstracts – starts with the comprehensive presentation of the abstract ecosystem, offering also a specific perspective on SSH. Subchapter 5.2 offers solutions to the issues of “missing abstracts” which are aimed at the needs of the GoTriple platform. Chapter 6 – dedicated to citations – offers an overview of the topic and its relevance to the SSH. In the subchapter 6.2 an analysis of issues related to GoTriple’s expression of citation data is presented (that relates especially to the challenge of processing different citation formats and citation data quality). Each chapter concludes with a summary of the guidelines for the specific metadata type for the humanities., The TRIPLE project (https://project.gotriple.eu/), which is financed under the Horizon 2020 framework (https://cordis.europa.eu/project/id/863420), under Grant Agreement No. 863420, with approx. 5.6 million Euros for a duration of 42 months (2019-2023). The content of this deliverable reflects only TRIPLE's view and the Commission is not responsible for any use that may be made of the information it contains.

Whereas the prefabricated status of idioms or restricted collocations is relatively self-evident in their context of use, the “underlying rigidity” (Sinclair 1991: 110) of other types of phraseological units may only become evident through large-scale analyses of reference corpora. This chapter focuses on the identification of subtle lexico-grammatical petrification of multiword units in dependency-annotated corpora. More specifically, it investigates restrictions on the valency of binary collocations and their tendency to be regularly subsumed by larger collocational chains. For example, the binary collocation deep breath is almost invariably a direct object of a small set of verbs: take, draw, let out. This restriction can be contrasted with collocational chains in which other adjectival collocations of breath (e.g. bad breath) have a wider range of syntactic roles determined mainly by the potential valency of their head noun (i.e. its propensity to function as subject, object etc.). Apart from discussing examples of such constructions from Polish and English corpus data, the chapter also attempts to show how lexico-syntactic properties of multiword units can be systematically accounted for and explored using a dependency-based approach to phraseology extraction.

This article presents the current outcomes of the MARCELL CEF Telecom project aiming to collect and deeply annotate a large comparable corpus of legal documents. The MARCELL corpus includes 7 monolingual sub-corpora (Bulgarian, Croatian, Hungarian, Polish, Romanian, Slovak and Slovenian) containing the total body of respective national legislative documents. These sub-corpora are automatically sentence split, tokenized, lemmatized and morphologically and syntactically annotated. The monolingual sub- corpora are complemented by a thematically related parallel corpus (Croatian-English). The metadata and the annotations are uniformly provided for each language specific sub-corpus. Besides the standard morphosyntactic analysis plus named entity and dependency and/or noun phrase annotation, the corpus is enriched with the IATE and EuroVoc labels. The file format is CoNLL-U Plus Format, containing the ten columns specific to the CoNLL-U format and four extra columns specific to our corpora. The MARCELL corpora represent a rich and valuable source for further studies and developments in machine learning, cross-lingual terminological data extraction and classification.

This article provides an overview of the dissemination work carried out in META-NET from 2010 until early 2014; we describe its impact on the regional, national and international level, mainly with regard to politics and the situation of fundingfor langauge technology topics. This paper documents the initiative’s work throughout Europe in order to boost progress and innovation in our field.

Publikacja finansowana ze środków na naukę w latach 2007-2012 w ramach projektu rozwojowego.

The paper intends to give a brief summary of one the most recent efforts on building the pan-European language technology infrastructure: META-NET – a network of Excellence consisting of 54 research centres from 33 countries – and specifically, its Central and South-European participating project: CESAR. One of the major activities of the project is selection of the resources and tools to be collected, validated, standardized, upgraded/extended/cross-lingually aligned and stored in the META-SHARE open resource exchange facility. The contribution focuses on presenting the repository maintaining the metadata of the selected resources, the methodology and criteria for their selection and a detailed view to the resources and tools delivered by the project in 2011. After highlighting the concepts of META-SHARE metadata model and synchronized network of metadata servers, the article presents the methodology and criteria for the resource selection by calculating point values basing on solid evaluation indicators such as resource availability, quality, and quantity of similar resources available, coverage, maturity, sustainability and adaptability. The META-NET Language White Papers – the series of reports on the state of each European language with respect to language technology is also presented as well as the licensing guidelines put forward by the META-SHARE community, promoting open and free of charge use of data and tools by using standardized and well-defined legal attributions.

Publikacja finansowana ze środków na naukę w latach 2007-2012 w ramach projektu rozwojowego.

Publikacja finansowana ze środków na naukę w latach 2007-2012 w ramach projekt rozwojowy. Publikacja finansowana ze środków na naukę w latach 2007-2012 w ramach projektu rozwojowego.

Rozdział z książki: Narodowy Korpus Języka Polskiego, red. A. Przepiórkowski, M. Bańko, R. L. Górski, B. Lewandowska-Tomaszczyk, Wydawnictwo Naukowe PWN, Warszawa 2012, ISBN 978-83-01-16700-4 Publikacja finansowana ze środków na naukę w latach 2007-2012 w ramach projektu rozwojowego.

This article presents the current outcomes of the CURLICAT CEF Telecom project, which aims to collect and deeply annotate a set of large corpora from selected domains. The CURLICAT corpus includes 7 monolingual corpora (Bulgarian, Croatian, Hungarian, Polish, Romanian, Slovak and Slovenian) containing selected samples from respective national corpora. These corpora are automatically tokenized, lemmatized and morphologically analysed and the named entities annotated. The annotations are uniformly provided for each language specific corpus while the common metadata schema is harmonised across the languages. Additionally, the corpora are annotated for IATE terms in all languages. The file format is CoNLL-U Plus format, containing the ten columns specific to the CoNLL-U format and three extra columns specific to our corpora as defined by Varádi et al. (2020). The CURLICAT corpora represent a rich and valuable source not just for training NMT models, but also for further studies and developments in machine learning, cross- lingual terminological data extraction and classification.