metadata
license: cc-by-sa-4.0
We introduce the Million Authors Corpus (MAC), a novel dataset encompassing contributions from dozens of languages on Wikipedia. It includes only long and contiguous textual chunks taken from Wikipedia edits and links those texts to their authors.