Linghub2: Language Resource Discovery Tool for Language Technologies

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 6352–6360 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 6352 Linghub2: Language Resource Discovery Tool for Language Technologies C´ecile Robin∗, Gautham Vadakkekara Suresh∗, V´ıctor Rodr´ıguez Doncel†, John McCrae∗, Paul Buitelaar∗ ∗Insight Centre for Data Analytics, Data Science Institute, National University of Ireland Galway Lower Dangan, Galway, Ireland †Polytechnic University of Madrid Avenida Ramiro de Maeztu, s/n, Madrid, Spain {cecile.robin, gautham.suresh, john.mccrae}@insight-centre.org paul.buitelaar@nuigalway.ie, victor.rodriguez@upm.es Abstract Language resources are an essential component of natural language processing, as well as related research and applications. Users of language resources have different needs in terms of format, language, topics, etc. for the data they need to use. Linghub (McCrae and Cimiano, 2015) was first developed for this purpose, using the capabilities of linked data to represent metadata, and tackling the heterogeneous metadata issue. Linghub is aimed at helping language resources and technology users to easily find and retrieve relevant data, and identify important information on access, topics, etc. This work describes a rejuvenation and modernisation of the 2015 platform into using a popular open source data management system, DSpace, as foundation. The new platform, Linghub2, contains updated and extended resources, more languages, and continues the work towards the homogenisation of metadata through conversions, through linkage to standardisation strategies and community groups, such as the Open Digital Rights Language (ODRL) community group. Keywords: linked data, data repository, language resources, linguistics, open digital repository 1. Introduction Discovering and searching for adequate Language Re- sources (LRs) (including services) is at the core of any Natural Language Processing (NLP) task. The impor- tance of these steps should not be underestimated, as they have a significant impact on the technologies built from them. Several portals exist, but do not provide so- lutions that takes into account data quality, data cover- age and data standardisation all at once. The omission of any of these aspects reduces the chance of the data being discovered and used by stakeholders. The Linghub (McCrae and Cimiano, 2015) platform was first developed in 2015 in response to a lack of the aforementioned aspects. It makes use of the capabili- ties of Linked open Data (LOD) to represent metadata, using standards so that users does not need to learn new formats. Linghub’s main objectives are to provide in- formation on data that is open access, traceable, that can be queried using the LD query language SPARQL, and that tackles the issue of heterogeneous metadata. In this new version of Linghub, Linghub21, we push this further, by creating an improved user experience with advanced and stable search capabilities either through the Web interface or SPARQL queries, by integrating the use of standards and of quality tests, and by pro- viding more data with previous repositories and new datasets all available under the same interface and de- scribed using the same schemas. The work described in this paper is a result of the H2020 Prˆet-`a-LLOD project2 - Ready-to-use Multilin- 1https://new.linghub.org/ 2https://pret-a-llod.github.io/ …

Content truncated to 500 words