Una eina per a una llengua en procés d'estandardització: el traductor automàtic català-sard

IF 0.3 Q4 LINGUISTICS

Linguamatica Pub Date : 2018-01-16 DOI:10.21814/lm.9.2.255

Gianfranco Fronteddu, Hèctor Alòs i Font, Francis M. Tyers

{"title":"Una eina per a una llengua en procés d'estandardització: el traductor automàtic català-sard","authors":"Gianfranco Fronteddu, Hèctor Alòs i Font, Francis M. Tyers","doi":"10.21814/lm.9.2.255","DOIUrl":null,"url":null,"abstract":"catalaAquest article presenta el desenvolupament d'un sistema de traduccio automatica en codi obert basat en regles del catala al sard mitjancant la plataforma Apertium, parant una atencio especial a la creacio del diccionari bilingue i de les regles de seleccio lexica i transferencia estructural. Es mostren alguns problemes derivats de l'estat actual del sard estandard. S'ha obtingut una tassa d'error per paraula (WER) del 20,5% i una tassa d'error per paraula independent de la posicio (PER) del 13,9%. Mitjancant l'analisi qualitativa de la traduccio de quatre articles enciclopedics, s'analitzen les causes d'aquests resultats. EnglishThis article describes the development of a free/open-source rule-based machine translation system for Catalan to Sardinian based on the Apertium platform. Special attention is given to the components of the system related with transfer (structural and lexical) and lexical selection, drawing attention to issues stemming from the current state of the Sardinian written norm. The system has a word-error rate (WER) of 20.5% and a position-independent word-error rate (PER) of 13.9%. We analyse the remaining errors by doing a qualitative analysis of the translation of four articles from the encyclopaedic domain.","PeriodicalId":41819,"journal":{"name":"Linguamatica","volume":null,"pages":null},"PeriodicalIF":0.3000,"publicationDate":"2018-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Linguamatica","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.21814/lm.9.2.255","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"LINGUISTICS","Score":null,"Total":0}

引用次数: 1

Abstract

catalaAquest article presenta el desenvolupament d'un sistema de traduccio automatica en codi obert basat en regles del catala al sard mitjancant la plataforma Apertium, parant una atencio especial a la creacio del diccionari bilingue i de les regles de seleccio lexica i transferencia estructural. Es mostren alguns problemes derivats de l'estat actual del sard estandard. S'ha obtingut una tassa d'error per paraula (WER) del 20,5% i una tassa d'error per paraula independent de la posicio (PER) del 13,9%. Mitjancant l'analisi qualitativa de la traduccio de quatre articles enciclopedics, s'analitzen les causes d'aquests resultats. EnglishThis article describes the development of a free/open-source rule-based machine translation system for Catalan to Sardinian based on the Apertium platform. Special attention is given to the components of the system related with transfer (structural and lexical) and lexical selection, drawing attention to issues stemming from the current state of the Sardinian written norm. The system has a word-error rate (WER) of 20.5% and a position-independent word-error rate (PER) of 13.9%. We analyse the remaining errors by doing a qualitative analysis of the translation of four articles from the encyclopaedic domain.

查看原文本刊更多论文

一种语言标准化过程中的工具：加泰罗尼亚语撒丁语自动翻译器

catala本文通过Apertium平台开发了一个基于规则的自动开源翻译系统，从catala到sard，特别关注双语词典的创建以及词汇选择和结构转换的规则。从标准sard的现状可以看出一些问题。已获得20.5%的每字错误率（WER）和13.9%的与位置无关的每字误差率（per）。通过对四篇百科全书式文章的翻译进行定性分析，分析造成这些结果的原因。英语特别关注与迁移（结构和词汇）和词汇选择相关的系统组成部分，引起人们对撒丁岛书面规范现状的关注。该系统的单词错误率（WER）为20.5%，位置无关错误率（PER）为13.9%。我们通过对百科全书领域的四篇文章的翻译进行定性分析来分析剩余的错误。

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

Linguamatica LINGUISTICS-

CiteScore

1.40

自引率

0.00%

发文量

审稿时长

6 weeks