📚 Institutional Books 1.0

Institutional Books is a growing corpus of public domain books. This 1.0 release is comprised of 983,004 public domain books digitized as part of Harvard Library's participation in the Google Books project and refined by the Institutional Data Initiative. Use of this data is governed by the IDI Terms of Use for Early-Access.

983K books, published largely in the 19th and 20th centuries 242B o200k_base tokens 386M pages of text, available in both original… See the full description on the dataset page: https://huggingface.co/datasets/institutional/institutional-books-1.0.

based on the dataset goodreadsbooks by jealousleopard. used this dataset and simply added genre information to it. I added genres to all but 97 books. note that I upload this dataset in 2022 but it is based on a dataset which has probably not be updated for a couple of years.

to add the genres I used PaulKlinger "Enhance-GoodReads-Export" script with a few bug fixes, and enabled searching goodreads.com/en for data the script did not find under goodreads.com . this covered all books except about 400, 300 of which I extracted the genre data using a call to 'https://www.googleapis.com/books/v1/volumes?q=isbn:' the remaining 97 I leave to the kind-hearted to fill.

Book ratings

This dataset has two files:

Books_rating.csv --> With information about books ratings made by users books_data.csv --> Metadata about the books, title, author, genre, etc.

It is intended as an input dataset to train a recommender system. It was obtained from this dataset of Amazon book reviews in Kaggle

This database contains information about books gathered with help of Google Books API. The database contains 7 different tables where 3 of them are only to relate the other tables together. Tables: Books contains 1062 records. Authors contains 1595 records. Categories 109 records. Metadata 37 records. MD5 (GBooks_2015-06-09.sql) = bfd09094d0e123e668b2e58332b1a98b

Arabic Books

  Dataset Summary

The arabic-books dataset contains 8,500 rows of text, each representing the full text of a single Arabic book. These texts were extracted using the arabic-large-nougat model, showcasing the model’s capabilities in Arabic OCR and text extraction. The dataset spans a total of 1.1 billion tokens, calculated using the GPT-4 tokenizer. This dataset is a testimony to the quality of the Arabic Nougat models and their effectiveness in extracting… See the full description on the dataset page: https://huggingface.co/datasets/MohamedRashad/arabic-books.

Dataset: "Harry Potter" and "The Lord of the Rings" Text Corpus

This curated dataset on Kaggle comprises the entire collection of texts from the "Harry Potter" series by J.K. Rowling and "The Lord of the Rings" series, including "The Hobbit" and "The Silmarillion" by J.R.R. Tolkien. It serves as a comprehensive text corpus for training language models, particularly suited for text generation and natural language processing tasks.

About the Dataset:

Content: The dataset includes the complete text corpus from the "Harry Potter" series (seven books) and "The Lord of the Rings" trilogy (including "The Hobbit" and "The Silmarillion"). These iconic fantasy novels feature rich narratives, diverse characters, and intricate world-building. Format: The texts are provided in plain text format, divided into separate files or sections corresponding to each book/chapter.

Usage:

The combined collection of "Harry Potter" and "The Lord of the Rings" texts within this dataset offers a wealth of literary content for training language models, particularly for tasks like text generation, language modeling, sentiment analysis, and more. The dataset's diverse narratives and unique writing styles contribute to building robust and contextually aware language models.

This dataset is about books. It has 2,617,384 rows. It features 4 columns: author, book publisher, and ISBN. It is 97% filled with non-null values.

Books Borrowed - E-Library Resources CSV Performance Indicator : LIB02-R Books Borrowed - E-Library Explore Preview Download

Reading books remains a popular pastime for U.S. adults, with ** percent of respondents to a 2021 survey saying that they had read a book in any format within the last year. Despite online media formats now being the preferred option for many consumers when it comes to television, music, and gaming, print books are by far the most popular format among readers in the United States. Whilst almost double the share of adults now read audiobooks compared to 2011, only ** percent claimed to have read an audiobook in the last year compared to ** percent who said that they had read a print book. Book sales in the United States In 2020, bookstore sales in the United States amounted to **** billion U.S. dollars. Sales in 2019 and 2020 were the lowest recorded since the early *****, and the combined effect of the coronavirus outbreak, along with the growing appeal of online purchasing, will likely mean that bookstore sales will continue to drop. Bookstores tend to see most success in August, December, and January, and sales revenue often surpasses *********** U.S. dollars in those months each year. That said, monthly retail sales of bookstores in the U.S. are notably lower overall than in previous years and were particularly poor in spring 2020 as a result of national shutdowns to stem the spread of COVID-19. Influence of COVID-19 on reading habits The coronavirus pandemic led to increased media consumption in general, but not only among avid video and music streaming fans. Data from a survey in March 2020 revealed that ** percent of Millennials read more books due to the COVID-19 outbreak, making consumers in this group the most likely to have done so compared to ** percent of the total survey sample. Meanwhile, ** percent of Boomers said that their reading habits had not changed.

Book-Crossing dataset mined by Cai-Nicolas Ziegler

Freely available for research use when acknowledged with the following reference (further details on the dataset are given in this publication):

• PDF

• Cai-Nicolas Ziegler, Sean M. McNee, Joseph A. Konstan, Georg Lausen; Proceedings of the 14th International World Wide Web Conference (WWW '05), May 10-14, 2005, Chiba, Japan. To appear.

Further information and the original dataset can be found at the original webpage.

Changes to the dataset:

• Location removed as it comes in different formats not in default (city, state, country).
• Transferred from ISO-8859-1 to UTF-8
• Manually fixed a few rows with incorrect number of columns

Note:

• out of 278859 users:
  • only 99053 rated at least 1 book
  • only 43385 rated at least 2 books.
  • only 12306 rated at least 10 books.
• out of 271379 books:
  • only 270171 are rated at least once.
  • only 124513 have at least 2 ratings.
  • only 17480 have at least 10 ratings.

Damaged Books

## Overview

Damaged Books is a dataset for object detection tasks - it contains Damages annotations for 620 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

This dataset is about books. It has 132 rows and is filtered where the book publisher is The authors. It features 7 columns including author, publication date, language, and book publisher.

All Books

## Overview

All Books is a dataset for object detection tasks - it contains Book annotations for 2,070 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Books

## Overview

Books is a dataset for object detection tasks - it contains Novel Scientific 12 16 annotations for 884 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

AlekseyKorshuk/fiction-books dataset hosted on Hugging Face and contributed by the HF Datasets community

This dataset is about books. It has 43 rows and is filtered where the book publisher is Hachette Books. It features 7 columns including author, publication date, language, and book publisher.

A dataset constructed from book images using an optical character reader (OCR), an object detector, and a layout analyzer for the autonomous extraction of image–text pairs.

According to a survey held in the United States between March and April 2020, 70 percent of respondents said that they read print books the most, with 39 percent of those consumers preferring their books to be new.

The study was conducted as the U.S. went into lockdown to prevent the spread of the coronavirus, however although the virus certainly affected media consumption in the United States, what did not change was consumers' book preferences. Print has always been the most popular book format in the U.S. and figures on increased media consumption during the pandemic showed that even Gen Z, a generation famed for loving digital, were the most likely to be reading books more than usual during the outbreak.

Book consumption in the U.S.

Whilst printed newspapers and magazines have struggled to survive as digital formats grow ever more prevalent and appealing, when it comes to books U.S. consumers still have a clear preference for print. Annual survey data consistently shows that U.S. adults are far more likely to have read a print book in the last year than a digital version thereof, and whilst the popularity of digital books has increased, print remains the favorite.

As far as book buying goes, whilst the number of print books sold in the U.S. fluctuates each year, the figures remain relatively stable. Although unit sales have not surpassed 700 million since 2010, the number came close in 2018 and yearly sales from 2015 to 2019 were higher than the amount recorded in 2004.

Books Cover Dataset 2 [Backup]

## Overview

Books Cover Dataset 2 [Backup] is a dataset for object detection tasks - it contains Books OS6z annotations for 520 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [Public Domain license](https://creativecommons.org/licenses/Public Domain).