Bridging the “gApp”: improving neural machine translation systems for multiword expression detection

IF 0.1 0 LANGUAGE & LINGUISTICS

Yearbook of Phraseology Pub Date : 2020-11-25 DOI:10.1515/phras-2020-0005

Carlos Manuel Hidalgo-Ternero, G. C. Pastor

引用次数: 2

Abstract

Abstract The present research introduces the tool gApp, a Python-based text preprocessing system for the automatic identification and conversion of discontinuous multiword expressions (MWEs) into their continuous form in order to enhance neural machine translation (NMT). To this end, an experiment with semi-fixed verb–noun idiomatic combinations (VNICs) will be carried out in order to evaluate to what extent gApp can optimise the performance of the two main free open-source NMT systems —Google Translate and DeepL— under the challenge of MWE discontinuity in the Spanish into English directionality. In the light of our promising results, the study concludes with suggestions on how to further optimise MWE-aware NMT systems.

查看原文本刊更多论文

桥接“gApp”：改进用于多词表达检测的神经机器翻译系统

摘要本文介绍了gApp工具，这是一个基于Python的文本预处理系统，用于自动识别不连续的多词表达式并将其转换为连续形式，以增强神经机器翻译（NMT）。为此，将进行一项半固定动词-名词惯用组合（VNIC）实验，以评估gApp在多大程度上可以优化两个主要的免费开源NMT系统——谷歌翻译和DeepL——在西班牙语到英语的MWE不连续性的挑战下的性能。鉴于我们有希望的结果，该研究最后提出了如何进一步优化MWE感知NMT系统的建议。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Yearbook of Phraseology LANGUAGE & LINGUISTICS-

CiteScore

0.80

自引率

28.60%

发文量

期刊介绍： The Yearbook of Phraseology is a fully international, peer-reviewed publication dedicated to research in phraseology, a linguistic subfield concerned with the study of word combinations of varying extent and type, and different degrees of fixedness. Word combinations are ubiquitous in language and constitute a significant resource for communication. Their study is of interest to many other subdisciplines of linguistics and even to other disciplines, throwing light on the make-up of constructions, their processing and learning, the make-up and modes of creation of complex building blocks of language, the methodology and use of corpora and statistical methods, as well as on the way in which language functions.