Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
json
Languages:
English
Size:
10K - 100K
License:
| license: mit | |
| task_categories: | |
| - text-classification | |
| language: | |
| - en | |
| tags: | |
| - compression | |
| - indices | |
| pretty_name: Uniquely identifies words with indexes | |
| # Dataset Card for Dataset Name | |
| This dataset consists of book titles with corresponding unique identifiers (UIDs) that can be used as labels. It was created to support projects requiring a standardized way of referencing books by both their titles and unique labels. The dataset is intended for use in classification tasks, document processing, and training AI models where accurate identification of books is necessary. | |
| 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 | |
| The book-title_BUID dataset is part of a system of datasets designed to train AI models in a multi-step process. It serves as the foundational dataset, teaching models to reference books using titles and UIDs. Following this, the sequential_marker_dataset trains the model on how to stitch and unstitch sections of text, while the stitch_context_dataset teaches the model how to draw contextual links between stitched sections. The combination of these datasets enables the model to learn how to handle complex document relationships and contexts. | |
| - **Curated by:** [Robert McNarland, McNarland Software Consultation Inc.] | |
| - **Funded by [optional]:** [Robert McNarland] | |
| - **Shared by [optional]:** [Robert McNarland] | |
| - **Language(s) (NLP):** [ English (titles from Project Gutenberg)] | |
| - **License:** [MIT] | |
| ### Dataset Sources [optional] | |
| - **Repository:** [R3troR0b/book-title_BUID] | |
| - **Paper [optional]:** [More Information Needed] | |
| - **Demo [optional]:** [More Information Needed] | |
| ## Uses | |
| ### Direct Use | |
| This dataset can be directly applied in: | |
| Document Classification: Assigning UIDs to books for document categorization. | |
| AI Model Training: Supporting models that need to identify books by both their title and a unique label. | |
| Part of a Larger Dataset System: This dataset is essential for training models in systems that handle text stitching and unstitching, using the sequential_marker_dataset and stitch_context_dataset. | |
| ### Out-of-Scope Use | |
| The dataset is not designed for uses unrelated to book identification or classification tasks and may not work effectively for non-English titles or broader multilingual tasks. | |
| ## Dataset Structure | |
| The dataset consists of: | |
| Fields: | |
| - text: The book title. | |
| - label: The unique identifier (UID) for each book. | |
| ## Dataset Creation | |
| ### Curation Rationale | |
| The dataset was created to support models in learning how to identify books using both titles and unique labels. It is a key component in a system that trains models to understand how to stitch and unstitch sections of text and derive contextual relationships between sections. | |
| ### Source Data | |
| #### Data Collection and Processing | |
| The book titles are sourced from Project Gutenberg and are exclusively in English. The dataset was generated programmatically using a custom-built application to extract and organize titles with their respective UIDs. | |
| #### Who are the source data producers? | |
| The dataset's titles are from Project Gutenberg, which provides a large collection of public domain books. The book titles were extracted by Robert McNarland's custom application. | |
| ### Annotations [optional] | |
| #### Annotation process | |
| There are no annotations in the dataset. The dataset contains book titles and their unique identifiers as-is, with no additional metadata or tags. | |
| #### Who are the annotators? | |
| <!-- This section describes the people or systems who created the annotations. --> | |
| [More Information Needed] | |
| #### Personal and Sensitive Information | |
| This dataset does not contain any personal, sensitive, or private information. | |
| ## Bias, Risks, and Limitations | |
| ### Recommendations | |
| - Limitations: The dataset is limited to English-language book titles and may not generalize to non-English contexts. It is intended for tasks involving document classification and text stitching/unstitching rather than general NLP tasks. | |
| - Biases: Titles are derived from Project Gutenberg's collection, which may reflect certain historical or cultural biases inherent in the source material. | |
| ## Citation [optional] | |
| [More Information Needed] | |
| **BibTeX:** | |
| [More Information Needed] | |
| **APA:** | |
| [More Information Needed] | |
| ## Glossary [optional] | |
| - Book UID (BUID): A unique identifier assigned to each book title in the dataset. | |
| - Sequential Marker Dataset: The next dataset in the system, used to teach models how to stitch and unstitch sections of text. | |
| - Stitch Context Dataset: A forthcoming dataset to teach models how to infer context between stitched sections of text. | |
| ## More Information [optional] | |
| [More Information Needed] | |
| ## Dataset Card Authors [optional] | |
| Robert McNarland, McNarland Software Consultation Inc. | |
| ## Dataset Card Contact | |
| robert.mcnarland@gmail.com |