Title Human performance in machine translation post-editing /
Translation of Title Žmogiškieji veiksniai mašininio vertimo postredagavimo procese.
Authors Vasiliauskienė, Sandra
Full Text Download
Pages 67
Keywords [eng] machine translation ; post-editing ; machine translation post-editing
Abstract [eng] The topic of this master final degree project is the human performance in machine translation postediting. The past decade has witnessed a notable evolution in machine translation (MT) technologies, which has highlighted the growing significance of MT and post-editing. Today, it is widely employed in both the industrial language business and the broader public. Advancements in technology coupled with the persistent demand for post-editing present translators with a noteworthy opportunity to enhance their skills and competencies. The assertion is made that translators ought to rely on technology to aid them in their professional duties, as it facilitates heightened levels of productivity and efficiency. The role of human translators has undergone significant transformation, as they are increasingly assuming the role of post-editors for machine translation. The performance, competencies, and abilities of post-editors are often a topic of discussion in both scientific and industrial contexts. Diverse projects necessitate distinct specifications. The efficacy of a post-edited text may vary depending on the nature of the project it is intended for. The duration required for proficiently postediting a text is subject to significant variation based on the project specifications, despite the consistency of the source material, unedited machine-translated text, and the human post-editor. The aim of the present investigation is to evaluate the efforts of post-editing in machine translation postediting procedures. In order to achieve this goal, three specific objectives were established: to provide a comprehensive review of the existing literature on the quality of machine translation output and the performance of post-editing; analyse the user activity data with a focus on temporal and cognitive efforts; indentify the primary issues encountered during post-editing of two distinct types of content texts; compare to determine the post-editing strategies based on Blain et al’s typology (2011) utilized in post-editing procedures.
Dissertation Institution Kauno technologijos universitetas.
Type Master thesis
Language English
Publication date 2023