Comparative Analysis of Translation Accuracy Between Swiftkey and Baidu Input Method Software Chinese-Indonesian Translation Function

Juliana Juliana, Stefanny Lunila

Abstract


With the rapid development of China and technology, SwiftKey and Baidu have introduced input methods with auxiliary translation functions to overcome language barriers between different countries. These two software applications provide real-time assistance for users facing language communication difficulties. This study examines 190 sentences from the websites of 12 Chinese companies operating in Indonesia, analyzing the translation techniques used by both input methods and the types of translation errors they produce. The results indicate that five main translation techniques are employed: literal translation, free translation, transliteration, addition, and omission. Additionally, seven types of translation errors were identified, including verb mistranslation, English abbreviation mistranslation, noun mistranslation, lexical omission, untranslated words, adjective mistranslation, and lexical substitution errors. Data analysis shows that the translation accuracy of the SwiftKey input method is 69%, while that of the Baidu input method is 31%. This indicates that SwiftKey's translation quality and accuracy surpass Baidu's by 38%. Therefore, compared to the Baidu input method, SwiftKey provides a more reliable translation quality.

Keywords


translation machine; comparative analysis; Baidu input method; SwiftKey input method

Full Text:

PDF

Refbacks

  • There are currently no refbacks.