---
dataset_info:
  features:
  - name: Paragraph Text
    dtype: string
  - name: cl100k_base
    sequence: int64
  splits:
  - name: train
    num_bytes: 782208667
    num_examples: 286370
  - name: validation
    num_bytes: 145918407
    num_examples: 53694
  - name: test
    num_bytes: 48906749
    num_examples: 17899
  download_size: 408407721
  dataset_size: 977033823
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---

# Dataset Card for Guttenberg Austrlia Collection

<!-- Provide a quick summary of the dataset. -->

This dataset card aims to be a base template for new datasets. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/datasetcard_template.md?plain=1).

## Dataset Details

### Dataset Description

<!-- Provide a longer summary of what this dataset is. -->



- **Curated by:** Matthew Altenburg
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [Custom license](https://gutenberg.net.au/licence.html).

### Dataset Sources [optional]

<!-- Provide the basic links for the dataset. -->

- **Repository:** [Gutenberg-Data](https://github.com/southern-cross-ai/Gutenberg-Data)
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]

## Uses

<!-- Address questions around how the dataset is intended to be used. -->


### Direct Use

<!-- This section describes suitable use cases for the dataset. -->
1) **Commercial Use with Acknowledgment:** While commercial use is permitted, it is recommended that users acknowledge Project Gutenberg Australia as the source. This should be emphasized in the "Direct Use" section, especially for commercial applications.
2) **Research and Educational Use:** The dataset can be freely used for research, academic, and educational purposes. These uses align well with the public domain status and should be encouraged in the "Direct Use" section.

### Out-of-Scope Use

<!-- This section addresses misuse, malicious use, and uses that the dataset will not work well for. -->
1) **Jurisdictional Compliance:** Since the works are only guaranteed to be in the public domain in Australia, it's crucial to mention that users outside Australia need to ensure they are not violating copyright laws in their own countries. This means that out-of-scope use includes any activities that might infringe on copyright in jurisdictions where the works are not in the public domain.
2) **Misuse or Malicious Use:** The public domain status does not inherently prevent misuse, but the license does request responsible use of the dataset. This includes avoiding the creation of harmful or deceptive content.
Respect for Source Attribution: While not legally required, the license suggests acknowledging Project Gutenberg Australia. Therefore, using the dataset without appropriate attribution, especially in a commercial context, could be considered out-of-scope or at least discouraged.

## Dataset Structure

<!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
This dataset consists of a single column called "paragraphs." Each entry in this column represents a paragraph extracted from various books and novels available on Project Gutenberg Australia. The dataset does not contain additional metadata such as author names, publication dates, or book titles, focusing purely on the textual content.
[More Information Needed]

## Dataset Creation

### Curation Rationale

<!-- Motivation for the creation of this dataset. -->
The primary motivation for creating this dataset was to collect a large corpus of Australian literature and other written works to support the development of an Australian Large Language Model (LLM). By focusing on texts that are culturally and historically significant to Australia, this dataset aims to enhance the performance and relevance of language models trained specifically for Australian English and the cultural context of Australia. The dataset provides a rich source of language data that can be used to train, fine-tune, or evaluate LLMs with a focus on Australian literature, expressions, and nuances.
[More Information Needed]

### Source Data

<!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->
The source data for this dataset comes from [Project Gutenberg Australia](https://gutenberg.net.au/), a repository of texts that are in the public domain within Australia. The data was collected using a web scraping tool called Scrapy, which was employed to systematically extract paragraphs from a wide range of Australian novels and books available on the Project Gutenberg Australia website. The data primarily consists of literary texts, including classic and contemporary Australian literature, which are freely available and representative of the Australian cultural and linguistic landscape.

This dataset serves as a foundational resource for developing AI models that better understand and generate Australian English, contributing to the broader goal of building robust, culturally aware AI systems.
#### Data Collection and Processing

<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->
The data was collected using a web scraping tool called Scrapy to extract paragraphs from various texts available on Project Gutenberg Australia. The selection criteria for the data focused on texts that are in the public domain within Australia, primarily Australian literature and other culturally significant works. The scraping process involved identifying and extracting paragraphs from these texts, which were then compiled into a single-column format in the dataset.

**Filtering and Normalization:** The dataset did not undergo extensive filtering or normalization during the initial collection process. However, basic cleaning was performed to remove extraneous HTML tags and other non-text elements that may have been present in the raw data. The paragraphs were extracted in their original form, preserving the language, spelling, and grammar used in the source texts.

**Tools and Libraries Used:** The primary tool used for data collection was Scrapy, a popular Python library for web scraping.

#### Who are the source data producers?

<!-- This section describes the people or systems who originally created the data. It should also include self-reported demographic or identity information for the source data creators if this information is available. -->
The [source data producers](https://gutenberg.net.au/background.html) are the original authors and publishers of the texts available on Project Gutenberg Australia. These include a range of Australian writers, both classic and contemporary, whose works have entered the public domain in Australia. Since the texts are historical and literary in nature, the demographic information about the original authors may vary widely, covering different periods, genres, and perspectives within Australian literature.

In many cases, the texts reflect the cultural and societal contexts of their time, providing insights into the identity, values, and experiences of their authors. However, specific demographic or identity information about the authors is generally not included within the dataset itself, as the focus is on the textual content.

### Annotations [optional]

<!-- If the dataset contains annotations which are not part of the initial data collection, use this section to describe them. -->

#### Annotation process

<!-- This section describes the annotation process such as annotation tools used in the process, the amount of data annotated, annotation guidelines provided to the annotators, interannotator statistics, annotation validation, etc. -->

[More Information Needed]

#### Who are the annotators?

<!-- This section describes the people or systems who created the annotations. -->

[More Information Needed]

#### Personal and Sensitive Information

<!-- State whether the dataset contains data that might be considered personal, sensitive, or private (e.g., data that reveals addresses, uniquely identifiable names or aliases, racial or ethnic origins, sexual orientations, religious beliefs, political opinions, financial or health data, etc.). If efforts were made to anonymize the data, describe the anonymization process. -->

[More Information Needed]

## Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

While the dataset is valuable for training models on Australian English, users should be aware of potential biases. The texts were selected based on their availability in the public domain, which may result in a skewed representation of Australian literature, favoring older, more established works. This could lead to an overemphasis on historical perspectives and underrepresentation of contemporary or marginalized voices.

Technical Limitations: The dataset consists of raw paragraphs without contextual information, which may limit its effectiveness in certain NLP tasks that require understanding of larger textual structures or metadata. Additionally, the lack of annotations or labels means that the dataset may require significant preprocessing before it can be used in specific applications.

### Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Given the potential biases and limitations, users should consider complementing this dataset with more diverse and contemporary sources to ensure a balanced representation of Australian literature. Care should be taken when using the dataset for tasks that require sensitivity to cultural or historical context, as the texts may include outdated or biased language.

For technical applications, it may be necessary to enrich the dataset with additional metadata or annotations, depending on the specific use case. Users are encouraged to critically evaluate the dataset’s suitability for their intended applications, particularly if the model’s outputs will be used in sensitive or impactful contexts.

## Citation [optional]

<!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->

**BibTeX:**

[More Information Needed]

**APA:**

[More Information Needed]

## Glossary [optional]

<!-- If relevant, include terms and calculations in this section that can help readers understand the dataset or dataset card. -->

[More Information Needed]

## More Information [optional]

[More Information Needed]

## Dataset Card Authors [optional]

[More Information Needed]

## Dataset Card Contact
For any further details conatct Matthew Altenburg: Matthew.Altenburg@anu.edu.au