jennifee's picture
Update README.md
51afae0 verified
metadata
license: mit
language:
  - en
pretty_name: Book Review Text Data

Dataset Card for Book Text Data

This dataset provides text-based reviews for fiction and nonfiction books.

Dataset Details

Dataset Description

For a selection of books on my bookshelf, I collected some text data. I selected 15 fiction and 15 nonfiction books. I then wrote three reviews for each book to create the first 90 examples, and then I wrote 5 hypothetical fiction book reviews and 5 hypothetical nonfiction book reviews. These reviews were collected for the 30 books + 10 hypotheticals that make up my original split, and then they were augmented to create 1600 additional examples.

  • Curated by: Jennifer Evans
  • Language(s) (NLP): English
  • License: MIT License

Uses

Direct Use

Intended use: evaluating if a book is fiction or nonfiction based on the review.

Out-of-Scope Use

This dataset could be used for other book-related evaluations, like how people describe books and what their book preferences are. It could also be used to evaluate subgenres of fiction and nonfiction, such as sci-fi versus fantasy.

Dataset Structure

  • dataset_info:

  • features:

name: Review

dtype: string

name: Fiction?

dtype: int64
  • splits:

name: original

num_bytes: 15506

num_examples: 100

name: augmented

num_bytes: 247944

num_examples: 1600

download_size: 52247

dataset_size: 263450

  • configs:

config_name: default

data_files:

split: original

path: data/original-*

split: augmented

path: data/augmented-*

Dataset Creation

Curation Rationale

As an avid reader, I was drawn to reviewing some of the recent books I've read. I was especially interested in evaluating how my reviews may differ from the professional development style books I read (nonfiction) from the sci-fi and fantasy books I read (fiction).

Source Data

Books were selected from my bookshelf. I aimed to pull a variety of books, spanning different genres and series.

Data Collection and Processing

I selected 30 representative books from my bookshelf and then wrote 100 original reviews, including 3 reviews for each book. This data was then augmented with NLTK methods like synonym swapping to create the augmented dataset.

Who are the source data producers?

Jennifer Evans wrote the reviews for each book. She reads a variety of fiction and nonfiction books, mainly on professional development, wellness, sci-fi, and fantasy.