Datasets:
File size: 2,700 Bytes
a8ec509 2e305e8 7bc8f99 a8ec509 f9e6181 a8ec509 f9e6181 a8ec509 f9e6181 a8ec509 |
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 |
---
license: cc-by-2.0
size_categories:
- n<1K
task_categories:
- translation
dataset_info:
features:
- name: mizo
dtype: string
- name: english
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 64215
num_examples: 758
- name: test
num_bytes: 16213
num_examples: 190
download_size: 63295
dataset_size: 80428
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
---
# Mizo-English Parallel Sentences
<!-- Provide a quick summary of the dataset. -->
This dataset consists of parallel sentences in Mizo and English, extracted from the book "Mizo Songs and Folk Tales". It is designed to facilitate language translation models and linguistic research comparing the structure and vocabulary of the Mizo language with English.
## Dataset Details
### Dataset Description
Total number of lines: 950<br>
Language(s) (NLP): English, Mizo<br>
License: [CC-BY-2.0](https://creativecommons.org/licenses/by/2.0/)<br>
### Direct Use
This dataset is primarily intended for training and evaluating machine translation models between Mizo and English. It may also be useful for studies in comparative linguistics, language teaching, and other NLP applications such as automatic language identification.
## Dataset Structure
### Load the dataset
Using Huggingface:
```python
from datasets import load_dataset
ds = load_dataset("proadhikary/parallel-english-mizo")
```
Using Pandas:
```python
import pandas as pd
splits = {'train': 'data/train-00000-of-00001.parquet', 'test': 'data/test-00000-of-00001.parquet'}
df = pd.read_parquet("hf://datasets/proadhikary/parallel-english-mizo/" + splits["train"])
```
### Data Splits
- `train`: Contains the majority of the parallel sentences for model training.
- `test`: 20% sentences reserved for testing model performance.
## Dataset Creation
### Curation Rationale
The dataset was curated to provide a resource for bilingual education and to support research in machine translation systems, aiming to improve linguistic accessibility between Mizo and English.
#### Initial Data Collection and Normalization
Sentences were extracted verbatim from the book "Mizo Songs and Folk Tales" and formatted into parallel text files. Text normalization procedures were applied to ensure consistency in punctuation and capitalization.
#### Data Collection and Processing
The dataset was manually reviewed to correct any inconsistencies and ensure high-quality translations. Special attention was given to maintaining the cultural and contextual integrity of the translations.
Cite before using, thanks! :) |