The Arabic Parallel Gender Corpus 2.0: Extensions and Analyses

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 174–180 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 174 Italian NLP for Everyone: Resources and Models from EVALITA to the European Language Grid Valerio Basile∗, Cristina Bosco∗, Michael Fell∗, Viviana Patti∗, Rossella Varvara† ∗University of Turin, Italy {valerio.basile,cristina.bosco,michaelkurt.fell,viviana.patti}@unito.it †University of Fribourg, Switzerland rossella.varvara@unifr.ch Abstract The European Language Grid enables researchers and practitioners to easily distribute and use NLP resources and models, such as corpora and classifiers. We describe in this paper how, during the course of our EVALITA4ELG project, we have integrated datasets and systems for the Italian language. We show how easy it is to use the integrated systems, and demonstrate in case studies how seamless the application of the platform is, providing Italian NLP for everyone. Keywords: Italian, European Language Grid, EVALITA 1. The EVALITA4ELG Project In the past 20 years the Natural Language Processing (NLP) area has been featured by a growing interest for standard formats for linguistic resources and for initia- tives devoted to improve the accessibility and sharing of them. While the former have led to large and in- ternational projects, e.g., Universal Dependencies for morpho-syntactic annotation, the latter resulted in a wider availability and usability of resources on plat- forms such as META SHARE1, European Language Grid (ELG)2, CLARIN3, and DARIAH4. In this paper, we describe a recent initiative developed within the context of ELG, namely the EVALITA4ELG project, and we show how easy it is to use the systems and resources integrated on ELG for several NLP tasks. ELG, launched in early 2021 and funded by the Eu- ropean Union’s Horizon 2020 research and innovation programme, is a large technological infrastructure aim- ing to become the primary platform for Language Tech- nology in Europe. As a pilot funded by ELG itself, the EVALITA4ELG project aimed at systematically collect the resources and systems released for the EVALITA evaluation campaign, and make them easily accessible through the ELG platform. EVALITA5 is an evaluation campaign centered on a va- riety of phenomena and tasks for the Italian language. Started in 2007, it has been organized every two years ever since, with the contribution of the national and in- ternational community working on NLP for the Italian language. Similar campaigns are supported by other communities working on other languages, such as the Workshop on Evaluation of Human Language Tech- 1http://www.meta-share.org/ 2https://www.european-language-grid. eu/ 3https://www.clarin.eu/ 4https://www.dariah.eu/ 5https://www.evalita.it/ nologies for modern Iberian languages6 (IberEval), which started in 2010, but has only been held on a reg- ular annual basis since 2017, and evolved into Iberian Languages Evaluation Forum (IberLEF); the Natural Language Processing shared tasks for German7 (Ger- mEval), which started in 2014; the D´efi Fouille de Textes8 (DEFT) for French, which started in 2005 as an annual event. Considering the diffusion of evaluation campaigns centered on other national languages, the expected impact of EVALITA4ELG is not only on the community focused on Italian, but also on other com- munities, for …

Content truncated to 500 words