An analysis on Google Translate accuracy in Chinese-Indonesian text translation  

Oleh:

Nan, Guan

Jenis: Article from Proceeding
Dalam koleksi: KOLITA 17: Konferensi Linguistik Tahunan Atma Jaya Ketujuh Belas Tingkat Internasional, page 166-169.
Topik: accuracy; Google Translate; machine translation (MT); text typology
Fulltext: 166-169.Guan Nan.pdf (428.92KB)

Ketersediaan

Perpustakaan PKBB Nomor Panggil: 406 KLA 17 Non-tandon: tidak ada Tandon: 1

Lihat Detail Induk

Isi artikel

In this globalization era, the continuous deepening of communications, exchanges, and cooperations among countries have led to a great demand for large amount of translation volume, which can no longer fulfilled only by human translation (HT). Since the emergence of machine translation (MT), its extremely fast speed of translation and the decreasing cost have made it become more and more popular in daily use. Many computational linguists have been working hard to improve the accuracy of MT to make it more applicable and practical. With the rapid development of internet technology, MT has played an increasingly important role in enhancing communication in various fields. This research named An Analysis on Google Translate Accuracy in Chinese - Indonesian Text Translation focuses on MT represented by Google Translate, finding the translation errors made by Google Translate from Chinese to Indonesian language based on different text types. The text types in this research mainly include the informative texts, the expressive texts, and the operative texts mentioned in Reiss's text typology. The linguistic data basically has covered the typical text types, such as news reports, literary works, speeches, text books, manuals, and advertisements. The MT results are directly translated using Google Translate, while the HT results of the three text types as comparisons and references are selected from official versions or recognized versions by professionals. The error quantity of the MT versions is counted to analyze the translation effect, in terms of four error types, including omission, addition, different meaning, and zero meaning. As a study of MT from the perspective of linguistics, this research focuses on the translation effects of MT on different types of text. The results will help people to use MT properly according to corresponding text types and the targeted usage of translation versions. Also some suggestions on the specialized development direction for MT will be provided.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali