Inferring translation candidates for multilingual dictionary generation with multi-way neural machine translation

• Mihael Arcan ^[1] ; Daniel Torregrosa ^[1] ; Sina Ahmadi ^[1] ; John P. McCrae ^[1]
  1. [1] National University of Ireland

     National University of Ireland

     Irlanda

     
• Localización: Proceedings of TIAD-2019 Shared Task – Translation Inference Across Dictionaries co-located with the 2nd Language, Data and Knowledge Conference (LDK 2019): Leipzig, Germany, May 20, 2019 / Jorge Gracia (ed. lit.), Besim Kabashi (ed. lit.), Ilan Kernerman (ed. lit.), 2019, págs. 13-23
• Idioma: inglés
• Enlaces
  • Texto completo (pdf)
• Resumen

  • In the widely-connected digital world, multilingual lexical resources are one of the most important resources, for natural language processing applications, including information retrieval, question answer- ing or knowledge management. These applications benefit from the multi- lingual knowledge as well as from the semantic relation between the words documented in these resources. Since multilingual dictionary creation and curation is a time-consuming task, we explored the use of multi-way neural machine translation trained on corpora of languages from the same family and trained additionally with a relatively small human-validated dictionary to infer new translation candidates. Our results showed not only that new dictionary entries can be identified and extracted from the translation model, but also that the expected precision and recall of the resulting dictionary can be adjusted by using different thresholds.