Skip to main content
zenodoopen

WAC Corpus - Wikipedia Abusive Conversations

<p>This repository contains conversations between Wikipedia editors, which are annotated in terms of various types of abuse, at the level of messages. This corpus is described in the following publication:</p> <p>N. C&eacute;cillon, V. Labatut, R. Dufour, and G. Linar&egrave;s, &ldquo;WAC: A Corpus of Wikipedia Conversations for Online Abuse Detection,&rdquo; in <em>12th Language Resources and Evaluation Conference</em>, 2020, pp. 1375&ndash;1383.&nbsp;⟨<a href="https://hal.archives-ouvertes.fr/hal-02497514">hal-02497514</a>⟩</p> <p>The repository also contains the figures shown in this article.</p> <p><strong>Sources. </strong>Our corpus aligns two existing corpora:</p> <ul> <li>Messages and conversation structures of&nbsp;<em>WikiConv</em> (<a href="https://github.com/conversationai/wikidetox/tree/master/wikiconv">https://github.com/conversationai/wikidetox/tree/master/wikiconv</a>)</li> <li>Manual annotations in toxicity of <em>Wikipedia Comment Corpus</em> (WCC -- <a href="https://doi.org/10.6084/m9.figshare.4054689">https://doi.org/10.6084/m9.figshare.4054689</a>)</li> </ul> <p><strong>Citation. </strong>If you use this dataset, please cite the above article.</p> <p><br><code>@InProceedings{Cecillon2020,</code><br><code>&nbsp; author &nbsp; &nbsp;= {C&eacute;cillon, No&eacute; and Labatut, Vincent and Dufour, Richard and Linar&egrave;s, Georges},</code><br><code>&nbsp; title &nbsp; &nbsp; = {{WAC}: A Corpus of {W}ikipedia Conversations for Online Abuse Detection},</code><br><code>&nbsp; booktitle = {12\textsuperscript{th} Language Resources and Evaluation Conference},</code><br><code>&nbsp; year &nbsp; &nbsp; &nbsp;= {2020},</code><br><code>&nbsp; pages &nbsp; &nbsp; = {1375-1383},</code><br><code>&nbsp; address &nbsp; = {Marseille, FR},</code><br><code>&nbsp; url&nbsp; &nbsp; &nbsp; &nbsp;= {http://www.lrec-conf.org/proceedings/lrec2020/pdf/2020.lrec-1.172.pdf},</code><br><code>}</code></p> <p>&nbsp;</p>

ShareScore

32/100

Overall dataset sharing score

Score breakdown

These five areas show where the dataset supports — or may limit — practical reuse.

Stewardship
8
Harmonization
4
Access
16
Reuse readiness
0
Engagement
4

Topics