Accuracy and acceptability of DeepL Translate in translating legal document

Zalikha, Zalikha (2024) Accuracy and acceptability of DeepL Translate in translating legal document. Sarjana thesis, UIN Sunan Gunung Djati Bandung.

Preview	Text (COVER) COVER.pdf Download (34kB) | Preview
Preview	Text (ABSTRAK) ABSTRACT-4.pdf Download (110kB) | Preview
Preview	Text (DAFTAR ISI) TABLE OF CONTENTS.pdf Download (95kB) | Preview
Preview	Text (BAB I) CHAPTER I (1).pdf Download (233kB) | Preview
	Text (BAB II) CHAPTER II-3.pdf Restricted to Registered users only Download (300kB) | Request a copy
	Text (BAB III) CHAPTER III-2.pdf Restricted to Registered users only Download (136kB) | Request a copy
	Text (BAB IV) CHAPTER IV-5.pdf Restricted to Registered users only Download (422kB) | Request a copy
	Text (BAB V) CHAPTER V-2.pdf Restricted to Registered users only Download (58kB) | Request a copy
	Text (DAFTAR PUSTAKA) REFERENCES-2.pdf Restricted to Registered users only Download (205kB) | Request a copy

Abstract

The rapid advancement of machine translation technologies, particularly exemplified by DeepL Translate, has generated significant interest in their application to various domains, including legal document translation. However, to effectively integrate these technologies into legal practice, it is crucial to assess both the accuracy and acceptability of the translations produced. This research aims to investigate the performance of DeepL Translate in translating legal documents, evaluating its ability to maintain the accuracy while also considering the subjective aspect of acceptability. Nababan’s (2012) theories on Translation Quality Assessment (TQA) is the theoretical basis that is used to discuss the problem of accuracy and acceptability. This research employed a qualitative method that seeks to interpret the translation results from the data. The 47 data of this research consists of a collection of words, phrases, and sentences in Indonesian language contained in a legal document of the type of Study Completion Reference. Those data were collected by ways of observing, extracting, filtering, and translating the data. Those collected data were then analyzed by evaluating the translations, categorising the data, conducting descriptive analysis, and finally interpreting the results of the analysis. Based on the findings and discussions, it was found that there are 25 occurrences (53.2%) of accurate translation, 18 occurrences (38.3%) of less accurate translation, and 4 occurrences (8.5%) of inaccurate translation. Meanwhile for the acceptability, it was found that there are 23 occurrences (49%) of acceptable translation, 18 occurrences (38.3%) of less acceptable translation, and 6 occurrences (12.7%) of unacceptable translation. In a nutshell, the data highlights that DeepL Translate has a generally positive outcome for the users in translating legal document, with a majority of translations falling within the accurate and acceptable category.

Actions (login required)

View Item