metadata

language:
  - id
license:
  - cc-by-nc-4.0

Indonesian Children Books Dataset

This dataset contains text extracted from Indonesian children's books. This dataset is still contains raw text directly extracted from books, therefore still considered as dirty and need to be preprocessed further.

Dataset Description

Dataset Statistics

Number of books: 2,740
Total pages: 165,245
Total number of tokens: 25,759,439
Number of unique tokens: 698,094
Average tokens per page: 155.89

Extraction Methods Distribution

ocr: 16,064 pages
pymupdf: 148,951 pages
pypdf: 230 pages

Dataset Fields

text: The full text content of the page
title: The book title
page_num: The page number
extraction_method: Method used to extract text (OCR or direct text extraction)

Source

The books are gathered from publicly available children's books and educational materials. The dataset follows the Creative Commons Attribution-NonCommercial 4.0 International License.

Data Processing

Books were collected from various sources
Text was extracted using multiple methods:
- Direct text extraction from PDFs
- OCR for scanned pages