metadata

license: agpl-3.0
task_categories:
  - text-classification
language:
  - en
  - fr
  - de
  - ru
  - sv
tags:
  - webdataset
  - wikipedia

Dataset Card for WikiCCC

WikiCCC is a set of Wikipedia-based clustered (labeled) comparable corpus for clustering comparable corpora. A detailed description can be found in the paper Creating Clustered Comparable Corpora from Wikipedia with Different Fuzziness Levels and Language Representativity by Anna Laskina, Eric Gaussier, Gaelle Calvary and accepted at the 17th Workshop on Building and Using Comparable Corpora (BUCC 2024).