Replication Data for: \"Images of the arXiv: reconfiguring large scientific image datasets\"

Code and data for interacting with the article and image metadata from papers stored in the arXiv repository. At the time of writing the related publication, this involved investigating ~1.5 million articles, ~10 million images, and 2.1 TB of data downloaded from arXiv. This dataset upload contains instructions and code to download the bulk source data, extract into a folder hierarchy, create an SQLite database, and then run various queries, sample images, and generate plots using this data. The full SQLite database is provided containing article metadata, image metadata, and figure captions. Also contained here are data statistics and image credits for images that appear in the related publication.