Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -2,45 +2,12 @@
|
|
| 2 |
|
| 3 |
## Overview
|
| 4 |
|
| 5 |
-
The [Australian Corpus of English (ACE)](https://figshare.mq.edu.au/articles/dataset/Australian_Corpus_of_English_ACE_/24629712?file=43778418) corpus was compiled to match Australian data from 1986 to the standard American and British corpora (Brown and LOB) from the 1960s. It includes **1 million words of published text** in **500 samples from 15 categories of nonfiction and fiction**.
|
| 6 |
-
|
| 7 |
**Keywords**: Australian English, Corpus linguistics.
|
| 8 |
|
|
|
|
|
|
|
| 9 |
## Data Source
|
| 10 |
|
| 11 |
The original dataset is from [Macquarie University Research Data - Australian Corpus of English (ACE)](https://figshare.mq.edu.au/articles/dataset/Australian_Corpus_of_English_ACE_/24629712?file=43778418) and licensed under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).
|
| 12 |
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
After unzipping `ACE.zip`, the dataset under `ACE` contains:
|
| 16 |
-
|
| 17 |
-
- `data` contains 1718 `.txt` files including **plain** and **raw** two formats for each document.
|
| 18 |
-
- `Manul` contains 18 `.HTM` files with **source subtitles** and **author** information for each document.
|
| 19 |
-
|
| 20 |
-
## Download
|
| 21 |
-
|
| 22 |
-
You can download it directly from [Macquarie University Research Data - Australian Corpus of English (ACE)](https://figshare.mq.edu.au/articles/dataset/Australian_Corpus_of_English_ACE_/24629712?file=43778418).
|
| 23 |
-
|
| 24 |
-
You can also download it by running `download.py` in your terminal:
|
| 25 |
-
|
| 26 |
-
```bash
|
| 27 |
-
$ python3 download.py --help
|
| 28 |
-
usage: download.py [-h] [--save_path SAVE_PATH] [--unzip]
|
| 29 |
-
|
| 30 |
-
Download a file and optionally unzip it.
|
| 31 |
-
|
| 32 |
-
options:
|
| 33 |
-
-h, --help show this help message and exit
|
| 34 |
-
--save_path SAVE_PATH
|
| 35 |
-
Path to save the downloaded file.
|
| 36 |
-
--unzip Unzip the file if it's a zip archive.
|
| 37 |
-
```
|
| 38 |
-
|
| 39 |
-
For example:
|
| 40 |
-
|
| 41 |
-
- `python3 download.py --save_path my_data --unzip` will download and unzip the dataset `ACE.zip` under the directory `my_data`.
|
| 42 |
-
- `python3 download.py` will only download under the current directory.
|
| 43 |
-
|
| 44 |
-
## License
|
| 45 |
-
|
| 46 |
-
This repository is licensed under [MIT](https://opensource.org/license/mit).
|
|
|
|
| 2 |
|
| 3 |
## Overview
|
| 4 |
|
|
|
|
|
|
|
| 5 |
**Keywords**: Australian English, Corpus linguistics.
|
| 6 |
|
| 7 |
+
The [Australian Corpus of English (ACE)](https://figshare.mq.edu.au/articles/dataset/Australian_Corpus_of_English_ACE_/24629712?file=43778418) corpus was compiled to match Australian data from 1986 to the standard American and British corpora (Brown and LOB) from the 1960s. It includes **1 million words of published text** in **500 samples from 15 categories of nonfiction and fiction**.
|
| 8 |
+
|
| 9 |
## Data Source
|
| 10 |
|
| 11 |
The original dataset is from [Macquarie University Research Data - Australian Corpus of English (ACE)](https://figshare.mq.edu.au/articles/dataset/Australian_Corpus_of_English_ACE_/24629712?file=43778418) and licensed under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).
|
| 12 |
|
| 13 |
+
The current dataset is cleaned by [Gillian Law](https://huggingface.co/gillianalaw) and the uncleaned dataset can be found on [GitHub](https://github.com/southern-cross-ai/ACE).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|