Experiment on a phrase-based statistical machine translation using PoS Tag information for Sundanese into Indonesian

Abstract

This paper discusses the problem of Sundanese into Indonesian text translation, as one of low-resource language pair translation. The number of parallel corpus gives a significant impact on a statistical machine translation. Whereas to date, there are no Sundanese to Indonesian parallel corpus that ready to use. It is, therefore, we apply the PoS Tag rather than only surface form in the translation model to get a better translation result. This experiment was done to get an early result in Sundanese to Indonesian text translation and to identify problems arise on it. The result shows that the model using surface form and PoS Tag was slightly outperformed the model using only surface form. However, there are some problems faced in this experiment which are the large number of OOV caused by the limited number of parallel corpus and unproper phrase translation caused by some noise in the parallel corpus such as typos and inconsistency writing a word in Sundanese corpus.

Statistical Machine Translation Pada Bahasa Lampung Dialek Api Ke Bahasa Indonesia

“…Penerjemahan secara otomatis pada bahasa daerah dengan pendekatan DMT telah dilakukan oleh para peneliti. Beberapa penelitian tersebut diantaranya penelitian penerjemahan bahasa Indonesia ke aneka bahasa Sulawesi [4], penelitian penerjemahan bahasa Indonesia ke bahasa Jawa [5], [6], penelitian penerjemahan bahasa Indonesia ke bahasa Jawa menggunakan interpolation search [7], penelitian penerjemahan bahasa Bali ke bahasa Indonesia [8], penelitian penerjemahan bahasa Pali ke bahasa Sinhala menggunakan kamus dimana kedua bahasa tersebut ada di Sri Lanka [9], penelitian penerjemahan bahasa Kannada ke bahasa Telugu menggunakan kamus dimana kedua bahasa tersebut ada di India [10], penelitian penerjemahan secara otomatis bahasa daerah dengan menggunakan SMT telah dilakukan diantaranya oleh [5], [11]. Suryani, dkk melakukan penelitian penerjemahan secara statistik pada pasangan bahasa Sunda-Indonesia dengan menerapkan PoS Tag pada model translasinya.…”

Machine translation of Indonesian: a review

“…A. A. Suryani et.al [3] discussed Sundanese into Indonesian translation using phrase-based statistical machine translation with PoS Tag Information. They use three kinds of translation model.…”

“…This paper reviews the researches from a different angle of views: dataset, tools, method, and evaluation metric. Sometime, the research focuses on translation between Indonesian traditional language, such as [3]. The other research focuses on translation between Indonesian to foreign country language, such as [4].…”

“…The other research focuses on translation between Indonesian to foreign country language, such as [4]. The problem raises from lack of available dataset [3] with low resource of language pair. There are no Sundanese to Indonesian parallel corpus that ready to use.…”

Experiment on a Transformer Model Indonesian-to-Sundanese Neural Machine Translation with Sundanese Speech Level Evaluation

“…By now, we have not found yet any specific publications regarding the neural machine translation for Indonesian to Sundanese. However, from a previously done experiment by Suryani et al (2015) on Sundanese to Indonesian SMT, we tried to get the depiction of what Sundanese translation problems are, especially in semantic. The neural machine translation (NMT) is chosen as the current technology in machine translation and based on four big wins, according to WMT (World Conference on Machine Translation); this approach is more contextaware than other machine translation model.…”