holylovenia
commited on
Commit
•
f40f1b3
1
Parent(s):
a7d6988
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -3,4 +3,45 @@ tags:
|
|
3 |
- machine-translation
|
4 |
language:
|
5 |
- ind
|
6 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
- machine-translation
|
4 |
language:
|
5 |
- ind
|
6 |
+
---
|
7 |
+
|
8 |
+
"In the context of Machine Translation (MT) from-and-to English, Bahasa Indonesia has been considered a low-resource language,
|
9 |
+
and therefore applying Neural Machine Translation (NMT) which typically requires large training dataset proves to be problematic.
|
10 |
+
In this paper, we show otherwise by collecting large, publicly-available datasets from the Web, which we split into several domains: news, religion, general, and
|
11 |
+
conversation,to train and benchmark some variants of transformer-based NMT models across the domains.
|
12 |
+
We show using BLEU that our models perform well across them , outperform the baseline Statistical Machine Translation (SMT) models,
|
13 |
+
and perform comparably with Google Translate. Our datasets (with the standard split for training, validation, and testing), code, and models are available on https://github.com/gunnxx/indonesian-mt-data."
|
14 |
+
|
15 |
+
|
16 |
+
## Dataset Usage
|
17 |
+
|
18 |
+
Run `pip install nusacrowd` before loading the dataset through HuggingFace's `load_dataset`.
|
19 |
+
|
20 |
+
## Citation
|
21 |
+
|
22 |
+
```@inproceedings{guntara-etal-2020-benchmarking,
|
23 |
+
title = "Benchmarking Multidomain {E}nglish-{I}ndonesian Machine Translation",
|
24 |
+
author = "Guntara, Tri Wahyu and
|
25 |
+
Aji, Alham Fikri and
|
26 |
+
Prasojo, Radityo Eko",
|
27 |
+
booktitle = "Proceedings of the 13th Workshop on Building and Using Comparable Corpora",
|
28 |
+
month = may,
|
29 |
+
year = "2020",
|
30 |
+
address = "Marseille, France",
|
31 |
+
publisher = "European Language Resources Association",
|
32 |
+
url = "https://aclanthology.org/2020.bucc-1.6",
|
33 |
+
pages = "35--43",
|
34 |
+
language = "English",
|
35 |
+
ISBN = "979-10-95546-42-9",
|
36 |
+
}
|
37 |
+
```
|
38 |
+
|
39 |
+
## License
|
40 |
+
|
41 |
+
Creative Commons Attribution Share-Alike 4.0 International
|
42 |
+
|
43 |
+
## Homepage
|
44 |
+
|
45 |
+
### NusaCatalogue
|
46 |
+
|
47 |
+
For easy indexing and metadata: [https://indonlp.github.io/nusa-catalogue](https://indonlp.github.io/nusa-catalogue)
|