Abstract

Today, Machine Translation have an important role in communication. The need of Machine Translation System is getting higher in this information era. Some Machine Translation already exist, but many researcher interested to improve the quality of translation more natural. Find an optimal translation is not an easy thing to do in language processing. In this paper, we discuss about Machine Translation survey that contain Indonesia language to other language. There are different approaches to machine translation. Various method used in evaluating were also discussed like BLEU and NIST. Moreover, its future works to improve the translation quality. From the review results obtained that the translation has better performance depend on the number of corpus, well-behaved aligned corpus, and the technique used.

Highlights

  • Nowadays, machine translation has an important role in general communication

  • The attention-based approach is being increasingly used to improve the performance of neural machine translation (NMT)

  • Neural machine translation is widely used by researchers to the proposed translation system

Read more

Summary

Syntax-Based

The basic idea of syntax-based is the translation rule by synchronous grammar between source and target language. The rule for translation consists of sequence of words, syntax tree, and vector of feature value which describe the language pair. Warren Weaver had introduced the idea of Statistical Machine Translation [5]. SMT is one of the machine translation system using statistical approach which parameters are derived from the results of parallel corpus analysis. One weakness of SMT is the challenge of translating material that is not similar to content from the training corpora [8]. It gives poor accuracy of the translation result. SMT does not work well between languages that have significantly different word orders e.g. Japanese-Indonesian

Hierarchical-Phrase-Based
Hybrid Machine Translation
Hybrid with Multiple Approaches
Multi-Pass
Dataset
Sujaini 2014 English- 27K sentences
Future Work for Indonesian Machine Translation
Findings
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call