Dataset: - Navanjana/Gutenberg_books

Dataset Card for Gutenberg_books

Dataset Summary

The Gutenberg_books dataset comprises a collection of literary works from Project Gutenberg, formatted into a CSV file for ease of use in natural language processing (NLP) tasks. This dataset is intended for applications such as text classification, language modeling, and other NLP experiments.

Supported Tasks and Leaderboards

Language Modeling: Train models to predict the next word or sequence of words in a text.
Text Classification: Categorize texts into predefined genres or themes.
Text Generation: Generate human-like text based on the styles present in the dataset.

Languages

The dataset primarily includes works in English.

Dataset Structure

Data Instances

Each row in the CSV file represents a literary work with the following fields:

Title: The title of the work.
Text: The full text of the work.

Data Fields

Title (string): The title of the literary work.
Text (string): The complete text of the work.

Data Splits

The dataset is provided as a single CSV file without predefined splits. Users can create their own training, validation, and test splits as needed.

Dataset Creation

Curation Rationale

The dataset was curated to provide a diverse collection of literary works for NLP research and development, enabling tasks such as language modeling and text classification.

Source Data

Initial Data Collection and Processing

The texts were sourced from Project Gutenberg, a repository of free eBooks. Each text was processed to remove non-essential content like headers and footers, ensuring consistency across the dataset.

Who are the source data producers?

The original texts were digitized and provided by Project Gutenberg, a volunteer effort to archive cultural works.

Annotations

No additional annotations were added to this dataset.

Personal and Sensitive Information

The dataset contains public domain literary works and does not include personal or sensitive information.

Considerations for Using the Data

Social Impact of Dataset

This dataset enables the development of language models and other NLP tools that can understand and generate human-like text, contributing to advancements in AI literacy and accessibility.

Discussion of Biases

The dataset consists of literary works primarily from Western authors, which may introduce cultural or linguistic biases. Users should be aware of this limitation when training models.

Other Known Limitations

Some texts may contain outdated language or perspectives reflective of their time of writing.

Additional Information

Dataset Curators

The dataset was curated by Navanjana.

Licensing Information

The original texts are from Project Gutenberg and are in the public domain.

Citation Information

If you use this dataset, please cite Project Gutenberg as the source of the original texts.

Contributions

Contributions to this dataset are welcome. Please contact Navanjana for more information.