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

TL;DR

This paper demonstrates that abusive language detection models for online comments must account for language change over time, as static models degrade rapidly when applied to future data, highlighting the importance of temporal dynamics.

Contribution

It introduces a new German comments dataset and shows the importance of time-aware evaluation for abusive language detection models.

Findings

Naive ML models underperform on future data

Time stratified evaluation yields more realistic performance estimates

Model performance degrades rapidly over time when language evolves

Abstract

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