Assesing the Accuracy of Translation Result of Kataku Version 1.1 and Transtool 10 from English to Indonesian and Its Implication on Language Teaching

Abstract

This journal is aimed at investigating the accuracy of Kataku Version 1.1 and Transtool10 Rar and knowing its benefit for teaching writing. This is important because Computer Assisted Translation (CAT) has become a need and practical translation tool as a software to translate source language text (SL) into the target language (TL). The number of users of both devices requires a proof of whether both tools are qualified enough to translate from SL to TL. This study used descriptive qualitative methods with data collection techniques using documents, interviews and questionnaires. While the data analysis technique used component analysis consisting of four parts, that is, domain, taxonomy, component and cultural theme analysis using “criteria based sampling” 1 to 3, that is, accurate (3), less accurate (2) and not accurate (1). The results showed the quality of the "Kataku Version 1.1”was 24% accurate, 32% less accurate, and 45% inaccurate, while "Transtool. 10 Rar" was 33% accurate, 32% less accurate, and 35% inaccurate. Of the two CATs above, Transtool10 RAR has a higher level of accuracy than Kataku Version 1.1. In terms of error, the most common mistakes made by both were sequentially at semantic level, syntax, phrase, word order, lexical, lost in contact, and word content. However, from the error rate made by the two translation tools, Transtools 10 Rar got less error than the Kataku Version 1.1. In the case of teaching language, using these two software in translation was still very advantageous, especially for teaching writing and reading. By knowing the quality results of translation from one language to another, the students got much improvement by analyzing errors of vocabulary usages, grammar, the messages of the texts and writing products because as known that translator  was the  writer too. However, there was still a little obstacles using both software especially if the users saw the results of translation only as the collection of words which were isolated and independent. Its effects led to the misuses of the words either it is used in collocation or in  terms of different genre of the texts.

Thus it can be concluded that the two tools are not feasible to be used to translate various texts without involving professional translation experts by using appropriate proofreading and editing. In addition, for teaching language, both software are still effective and good for future applied language teaching and learning.