sbp / README.md
ben4mn's picture
Update README.md
46b317e verified
|
raw
history blame
1.06 kB
metadata
title: Dataset Name
tags:
  - text
  - retrieval
  - information-retrieval
  - question-answering
  - dataset
license: apache-2.0
datasets:
  - your-username/your-dataset-name

Dataset Name

Overview

A dataset created from a 90-page PDF document, structured for retrieval and information extraction tasks. Each entry contains textual content and metadata, making it ideal for applications such as question answering, summarization, and semantic search.

  • Source: Derived from a specific document (e.g., a book, article, or report).
  • Size: Contains individual entries for each page or section of the document.
  • Purpose: Designed for use in NLP retrieval and summarization tasks.

Dataset Structure

Each data entry includes:

  • content: The main text extracted from a page or section.
  • meta: Metadata associated with each entry. This can include page number, title, or other identifying information.

Example entry:

- content: "This is a sample page text."
  meta:
    page_number: 1
    title: "Sample Title"