Human Post-editing in Hybrid Machine Translation Systems: Automatic and Manual Analysis and Evaluation
Metadatos
Mostrar el registro completo del ítemEditorial
Springer International Publishing
Materia
Post-editing posedición hybrid machine translation traducción automática híbrida automatic evaluation human evaluation evaluación automática evaluación humana
Fecha
2018Referencia bibliográfica
Gutiérrez-Artacho, J., Olvera-Lobo, M.-D., & Rivera-Trigueros, I. (2018). Human Post-editing in Hybrid Machine Translation Systems: Automatic and Manual Analysis and Evaluation. En Trends and Advances in Information Systems and Technology (pp. 254-263). Springer International Publishing. DOI: https://doi.org/10.1007/978-3-319-77703-0_26
Resumen
This study assesses, automatically and manually, the performance of two hybrid machine translation (HMT) systems, via a text corpus of questions in the Spanish and English languages. The results show that human evaluation metrics are more reliable when evaluating HMT performance. Further, there is evidence that MT can streamline the translation process for specific types of texts, such as questions; however, it does not yet rival the quality of human translations, to which post-editing is key in this process.