English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The spread of online hate has become a significant problem for newspapersthat host comment sections. As a result, there is growing interest in usingmachine learning and natural language processing for (semi-) automated abusivelanguage detection to avoid manual comment moderation costs or having to shutdown comment sections altogether. However, much of the past work on abusivelanguage detection assumes that classifiers operate in a static languageenvironment, despite language and news being in a state of constant flux. Inthis paper, we show using a new German newspaper comments dataset that theclassifiers trained with naive ML techniques like a random-test train splitwill underperform on future data, and that a time stratified evaluation splitis more appropriate. We also show that classifier performance rapidly degradeswhen evaluated on data from a different period than the training data. Ourfindings suggest that it is necessary to consider the temporal dynamics oflanguage when developing an abusive language detection system or risk deployinga model that will quickly become defunct.

No Time Like the Present: Effects of Language Change on Automated Comment Moderation