ANALYSIS OF THE ACCURACY AND ACCEPTABILITY ON ABBREVIATION PHRASES TRANSLATION OF CNN INDONESIA INSTAGRAM POST CAPTION

The raising of technology leads Machine Translation become assisting tool to translate the text in the social media. For example, in Instagram, developed by Meta, the latest version called Neural Machine Translation work by using neural networks and builds and trains an extensive neural network that reads a sentence and outputs correct translation, even this translation is used by CNN Indonesia on Instagram. This study aims to identify, investigate, and examine the abbreviation phrases on the caption, the translation techniques on the abbreviation phrases, and the quality translation of the abbreviation phrases on CNN Indonesia Instagram. The data were collected through content analysis and focus group discussion. The researcher found 72 data contain of abbreviation types such as 51 data of abbreviation and 21 data of acronymization. The translation technique applied in are as Amplification (7%), Borrowing (53%), Discursive Creative (3%), Established Equivalent (1%), Established Equivalent & Borrowing (33%), and Reduction (2%). The application of the technique led to the sum of 2.50 for accuracy and 1.80 for acceptability on abbreviation, and 2.30 for accuracy and 2.00 for acceptability of acronymization.