---
license: mit
language:
- en
tags:
- australia
- corpus
- english
size_categories:
- 1M<n<10M
---
# Australian Corpus of English (ACE)	

## Overview

**Keywords**: Australian English, Corpus linguistics.

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**.

## Data Source

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/).

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).