Traducción Automática Interactiva Basada en Segmentos de Palabras

Current interactive machine translation systems are based on the validation / correction by an human of the successive prefixes of the translations and the generation of the corresponding suffixes by the machine translation systems. This approach has the disadvantage that sometimes one correction by...

ver descrição completa

Detalhes bibliográficos
Autor: Torres Badia, Guillem
Formato: tesis de maestría
Fecha de publicación:2016
País:España
Recursos:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:español
OAI Identifier:oai:riunet.upv.es:10251/76842
Acesso em linha:https://riunet.upv.es/handle/10251/76842
Access Level:acceso abierto
Palavra-chave:Machine translation
Hierarchical models
Traducción automática
Interactive machine translation
Hypergraphs
Traducción interactiva
Modelos jerárquicos
Hipergrafos
LENGUAJES Y SISTEMAS INFORMATICOS
Máster Universitario en Inteligencia Artificial, Reconocimiento de Formas e Imagen Digital-Màster Universitari en Intel·ligència Artificial, Reconeixement de Formes i Imatge Digital
Descrição
Resumo:Current interactive machine translation systems are based on the validation / correction by an human of the successive prefixes of the translations and the generation of the corresponding suffixes by the machine translation systems. This approach has the disadvantage that sometimes one correction by the human causes the system to generate a suffix of poorer quality than existed before correction. This work presents an efficient implementation of an interactive-predictive machine translation system in which the human validates all desired segments of the automatic generated translations and introduces a correction, the system must fill with new suggestions the segments not validated by the human. This implementation will be validated through a series of experiments in various translation tasks.