Human evaluation of MT
Identifiers
URI: http://hdl.handle.net/10317/3540Share
Metrics
Statistics
View Usage StatisticsMetadata
Show full item recordAuthor
Mendoza Mompeán, GinésDirector/a
Quesada Pereira, Fernando DanielCenter
Escuela Técnica Superior de Ingeniería de TelecomunicaciónUniversity
Universidad Politécnica de CartagenaDepartment
Tecnologías de la Información y las ComunicacionesKnowledge Area
Teoría de la Señal y las ComunicacionesPublication date
2013-10-23Keywords
Machine translationEvaluation
Applications
Human evaluation of MT
Abstract
[ENG]In this diploma work we developed a tool which can be used to evaluate the quality of machine translation systems, comparing the translations of MT with translations of native speaker. This application allows us to collect human judgment on translation output. For that to be possible we have implemented some tools such as 1) score the translation, 2) error classification, 3) fluency & adequacy of the translation, 4) mistranslated sentences. We have to say that the application has been developed to improve the machine translation system, although there are methods for automatic evaluation that reduce costs and runtime of evaluation process. Human evaluation is still necessary in MT research to improve these systems, since the results of automatic evaluation are not exact. [SPA]En este diploma hemos desarrollado una herramienta en la cual nosotros podremos evaluar la calidad de sistemas de traducción automática, comparando la traducción de un sistema de traducción con la traducción ...
Collections
The following license files are associated with this item:
Social media