File size: 6,200 Bytes
7ebb554
0ceedf1
7ebb554
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3dcfd2c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7ebb554
 
 
 
3dcfd2c
7ebb554
 
0ceedf1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7ebb554
0ceedf1
7ebb554
0ceedf1
3dcfd2c
 
7ebb554
0ceedf1
333e8f2
 
54e2680
7ebb554
 
 
ca504e4
7ebb554
67303e7
 
 
 
 
 
 
 
 
 
 
 
 
7ebb554
 
3dcfd2c
7ebb554
 
 
 
 
3dcfd2c
7ebb554
 
 
 
 
3dcfd2c
7ebb554
 
 
3dcfd2c
7ebb554
 
 
 
 
3dcfd2c
7ebb554
 
 
3dcfd2c
7ebb554
 
 
 
 
 
 
 
 
3dcfd2c
7ebb554
 
 
3dcfd2c
7ebb554
 
 
 
 
 
 
 
 
 
 
 
0ceedf1
a5642ac
0ceedf1
7ebb554
 
8a04a9b
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
# Dataset Card for allenai/wmt22_african

## Table of Contents
- [Dataset Description](#dataset-description)
  - [Dataset Summary](#dataset-summary)
  - [Supported Tasks](#supported-tasks-and-leaderboards)
  - [Languages](#languages)
- [Dataset Structure](#dataset-structure)
  - [Data Instances](#data-instances)
  - [Data Fields](#data-instances)
  - [Data Splits](#data-instances)
- [Dataset Creation](#dataset-creation)
  - [Curation Rationale](#curation-rationale)
  - [Source Data](#source-data)
  - [Annotations](#annotations)
  - [Personal and Sensitive Information](#personal-and-sensitive-information)
- [Considerations for Using the Data](#considerations-for-using-the-data)
  - [Social Impact of Dataset](#social-impact-of-dataset)
  - [Discussion of Biases](#discussion-of-biases)
  - [Other Known Limitations](#other-known-limitations)
- [Additional Information](#additional-information)
  - [Dataset Curators](#dataset-curators)
  - [Licensing Information](#licensing-information)
  - [Citation Information](#citation-information)

## Dataset Description

- **Homepage:** https://www.statmt.org/wmt22/large-scale-multilingual-translation-task.html
- **Repository:** [Needs More Information]
- **Paper:** [Needs More Information]
- **Leaderboard:** [Needs More Information]
- **Point of Contact:** [Needs More Information]

### Dataset Summary

This dataset was created based on [metadata](https://github.com/facebookresearch/LASER/tree/main/data/wmt22_african) for mined bitext released by Meta AI.  It contains bitext for 248 pairs for the African languages that are part of the [2022 WMT Shared Task on Large Scale Machine Translation Evaluation for African Languages](https://www.statmt.org/wmt22/large-scale-multilingual-translation-task.html).

#### How to use the data

There are two ways to access the data:

* Via the Hugging Face Python datasets library 

```
from datasets import load_dataset
dataset = load_dataset("allenai/wmt22_african")
```

* Clone the git repo
```
git lfs install
git clone https://huggingface.co/datasets/allenai/wmt22_african
```


### Supported Tasks and Leaderboards

This dataset is one of resources allowed under the Constrained Track for the [2022 WMT Shared Task on Large Scale Machine Translation Evaluation for African Languages](https://www.statmt.org/wmt22/large-scale-multilingual-translation-task.html).

### Languages

#### Focus languages

| Language | Code |
| -------- | ---- |
| Afrikaans | afr |
| Amharic | amh |
| Chichewa | nya |
| Nigerian Fulfulde | fuv |
| Hausa | hau |
| Igbo | ibo |
| Kamba | kam |
| Kinyarwanda | kin |
| Lingala | lin |
| Luganda | lug |
| Luo | luo |
| Northern Sotho | nso |
| Oroma | orm |
| Shona | sna |
| Somali | som |
| Swahili | swh |
| Swati | ssw |
| Tswana | tsn | 
| Umbundu | umb |
| Wolof | wol |
| Xhosa | xho |
| Xitsonga | tso |
| Yoruba | yor |
| Zulu | zul |

Colonial linguae francae: English - eng, French - fra

## Dataset Structure

The dataset contains gzipped tab delimited text files for each direction.  Each text file contains lines with parallel sentences.

### Data Instances

The dataset contains 248 language pairs.

Sentence counts for each pair can be found [here](https://huggingface.co/datasets/allenai/wmt22_african/blob/main/sentence_counts.txt).

### Data Fields

Every instance for a language pair contains the following fields: 'translation' (containing sentence pairs), 'laser_score', 'source_sentence_lid', 'target_sentence_lid', where 'lid' is language classification probability.

Example:
```
{
'translation': 
    {
        'afr': 'In Mei 2007, in ooreenstemming met die spesifikasies van die Java Gemeenskapproses, het Sun Java tegnologie geherlisensieer onder die GNU General Public License.', 
        'eng': 'As of May 2007, in compliance with the specifications of the Java Community Process, Sun relicensed most of its Java technologies under the GNU General Public License.'
    }, 
'laser_score': 1.0717015266418457, 
'source_sentence_lid': 0.9996600151062012, 
'target_sentence_lid': 0.9972000122070312
}
```
### Data Splits

The data is not split into train, dev, and test.

## Dataset Creation

### Curation Rationale

Parallel sentences from monolingual data in Common Crawl and ParaCrawl were identified via [Language-Agnostic Sentence Representation (LASER)](https://github.com/facebookresearch/LASER) encoders.

### Source Data

#### Initial Data Collection and Normalization

Monolingual data was obtained from Common Crawl and ParaCrawl.

#### Who are the source language producers?

Contributors to web text in Common Crawl and ParaCrawl.

### Annotations

#### Annotation process

The data was not human annotated. The metadata used to create the dataset can be found here: https://github.com/facebookresearch/LASER/tree/main/data/wmt22_african

#### Who are the annotators?

The data was not human annotated.  Parallel text from Common Crawl and Para Crawl monolingual data were identified automatically via [LASER](https://github.com/facebookresearch/LASER) encoders. 

### Personal and Sensitive Information

[Needs More Information]

## Considerations for Using the Data

### Social Impact of Dataset

This dataset provides data for training machine learning systems for many languages that have low resources available for NLP.

### Discussion of Biases

Biases in the data have not been studied.

### Other Known Limitations

[Needs More Information]

## Additional Information

### Dataset Curators

[Needs More Information]

### Licensing Information

The dataset is released under the terms of [ODC-BY](https://opendatacommons.org/licenses/by/1-0/). By using this, you are also bound by the Internet Archive [Terms of Use](https://archive.org/about/terms.php) in respect of the content contained in the dataset.

### Citation Information

NLLB Team et al, No Language Left Behind: Scaling Human-Centered Machine Translation, Arxiv, 2022.

### Contributions

We thank the AllenNLP team at AI2 for hosting and releasing this data, including [Akshita Bhagia](https://akshitab.github.io/) (for engineering efforts to create the huggingface dataset), and [Jesse Dodge](https://jessedodge.github.io/) (for organizing the connection).