Linking the TUFS Basic Vocabulary to the Open Multilingual Wordnet

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 307–316 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 307 Using Crowdsourced Exercises for Vocabulary Training to Expand ConceptNet Christos Rodosthenous1, Verena Lyding2, Federico Sangati3, Alexander König2,8 Umair ul Hassan4, Lionel Nicolas2, Jolita Horbacauskiene5, Anisia Katinskaia6 Lavinia Aparaschivei7 1 Computational Cognition Lab, Open University of Cyprus, Cyprus 2 Institute for Applied Linguistics, Eurac Research, Bolzano, Italy 3 Orientale University of Naples, Italy, 4 Insight Centre for Data Analytics, National University of Ireland Galway, Ireland 5 Kaunas University of Technology, Lithuania 6 Department of Computer Science, University of Helsinki, Finland 7 Faculty of Computer Science, Alexandru Ioan Cuza University of Iaşi, Romania 8 CLARIN ERIC, the Netherlands christos.rodosthenous@ouc.ac.cy, verena.lyding@eurac.edu, federico.sangati@gmail.com, alex@clarin.eu, umair.ulhassan@insight-centre.org, lionel.nicolas@eurac.edu, jolita.horbacauskiene@ktu.lt, anisia.katinskaia@helsinki.fi, aparaschiveilavinia96@gmail.com Abstract In this work, we report on a crowdsourcing experiment conducted using the V-TREL vocabulary trainer which is accessed via a Telegram chatbot interface to gather knowledge on word relations suitable for expanding ConceptNet. V-TREL is built on top of a generic architecture implementing the implicit crowdsourding paradigm in order to offer vocabulary training exercises generated from the commonsense knowledge-base ConceptNet and – in the background – to collect and evaluate the learners’ answers to extend ConceptNet with new words. In the experiment about 90 university students learning English at C1 level, based on the Common European Framework of Reference for Languages (CEFR), trained their vocabulary with V-TREL over a period of 16 calendar days. The experiment allowed to gather more than 12,000 answers from learners on different question types. In this paper, we present in detail the experimental setup and the outcome of the experiment, which indicates the potential of our approach for both crowdsourcing data as well as fostering vocabulary skills. Keywords: vocabulary trainer, commonsense knowledge, language learning 1. Introduction Language resources (LRs), like annotated corpora or dic- tionaries, are needed for many data-driven NLP tasks. Yet, refined resources of wide coverage are still lacking for many languages. This situation is mainly due to the fact that LRs are expensive to create and maintain. In order to address this persistent lack, we devised an ap- proach for facilitating the creation and extension of LRs by means of combining the domains of language learning and crowdsourcing and implemented a generic architecture for it (Rodosthenous et al., 2019). This implicit crowd- sourcing approach is unique in that it makes use of LRs to automatically generate language learning exercises, while in turn using learners’ answers to these exercises to extend or correct the underlying LR (cf. Section 5). Also, the application of the approach is supported by providing an open architecture which can easily be adapted and reused for different use-cases combining various LRs and language learning exercises. In this paper, we report on an experiment about vocabu- lary training of 81 university students learning English at C1 level, based on the Common European Framework of Reference for Languages (CEFR), by means of a Telegram chatbot named LingoGameBot, …

Content truncated to 500 words