meganariley/open-shelf

OpenShelf

A community-built, open book dataset. Readers browse, rate, and tag books through the OpenShelf Space — all contributions are written back here in real time.

Seed data comes from a personal Goodreads export (3,189 books, deduplicated). Community ratings and tags grow with each contribution.

Dataset Structure

books.parquet — 3,189 rows

The core book catalog. Seeded from Goodreads, enriched with community data over time.

Column	Type	Description
isbn13	str	ISBN-13 identifier
goodreads_id	str	Goodreads book ID
title	str	Book title
author	str	Primary author
additional_authors	str	Co-authors, if any
publisher	str	Publisher name
binding	str	Format (Hardcover, Paperback, etc.)
num_pages	Int32	Page count
year_published	Int32	Edition publication year
original_pub_year	Int32	Original publication year
goodreads_shelf	str	Shelf from the Goodreads export (read, to-read, currently-reading)
goodreads_rating	Int8	Seed rating from Goodreads (1–5), 0 if unrated
date_read	str	Date finished reading (YYYY/MM/DD), empty if not read
date_added	str	Date added to Goodreads
community_rating_count	Int32	Number of community ratings received
community_rating_sum	Int32	Sum of community ratings (divide by count for average)
genre_tags	str	Comma-separated genre tags from community
mood_tags	str	Comma-separated mood tags from community

Genre tags: Fiction, Non-Fiction, Mystery, Sci-Fi, Fantasy, Biography, History, Romance, Thriller, Literary, Essays, Poetry, Graphic Novel, Self-Help, Travel

Mood tags: Page-turner, Slow burn, Dense, Funny, Devastating, Uplifting, Unsettling, Cozy, Challenging, Breezy, Cerebral, Emotional

contributions.parquet — community-grown

One row per contribution. Written by the Space when a logged-in user rates or tags a book.

Column	Type	Description
contribution_id	str	UUID for this contribution
isbn13	str	ISBN-13 of the rated book
hf_username	str	HuggingFace username of the contributor
shelf	str	Shelf: read, to-read, currently-reading
rating	int8	Rating 1–5
genre_tags	str	Comma-separated genre tags applied
mood_tags	str	Comma-separated mood tags applied
contributed_at	str	ISO 8601 timestamp

Usage

import pandas as pd
from huggingface_hub import hf_hub_download

books = pd.read_parquet(
    hf_hub_download("meganariley/open-shelf", "books.parquet", repo_type="dataset")
)

# Community average rating (books with at least one rating)
rated = books[books["community_rating_count"] > 0].copy()
rated["avg_rating"] = rated["community_rating_sum"] / rated["community_rating_count"]
print(rated[["title", "author", "avg_rating"]].sort_values("avg_rating", ascending=False).head(10))

Data Source & License

Seed data exported from a personal Goodreads account. Community contributions are original and open.

Licensed under Open Data Commons Attribution License (ODC-By).

Related

• Space: meganariley/xet-shelf — browse, rate, and tag books