Neural Machine Translation (NMT): A New Technological Approache to Machine Translation (MT)

• Adina Rǎdulescu ^[1]
  1. [1] Spiru Haret University (Bucharest)
• Localización: "Un"-bordering Disciplinarity: "trans-/cross-/post-"disciplinary Approaches to Linguistic and Literary Research: (IInd Edition) / coord. por Denisa Drǎgușin, Daiana Dumbravescu, Ruxandra Vasilescu, 2019, ISBN 978-88-97908-51-7, págs. 211-219
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • The paper will focus on the latest research on the Neural Machine Translation domain and its advance as compared to the Statistical Machine Translation (SMT); it will also describe the intricate system and the steps that neural network translations go through to achieve the corresponding outputs in the target language. The literature on this topic has largely debated upon the advantages vs. disadvantages of NMT as compared to SMT, with a particular focus on the quality estimation of the output, as well as with regard to the human translator’s time and effort to post-edit the end products of NMT vs. SMT.