Algorithmic approaches to detecting, mitigating, and defending against online misogyny: a scoping review

Online misogyny remains a persistent and evolving problem across social media and digital platforms, manifesting through hate speech, harassment, and gender-based abuse. While early research has focused primarily on the automated detection of misogynistic content, there is growing recognition that algorithmic systems also play a role in mitigation, moderation, counter-speech, and broader defensive responses. Following consultation with external stakeholders, including practitioners and researchers working in violence prevention and digital governance, it became clear that a detection-only focus did not adequately capture the full range of algorithmic interventions currently being developed and deployed. In response, this scoping review broadens the analytical lens to examine how algorithmic methods are used not only to identify online misogyny, but also to mitigate its impacts and support defensive or preventative strategies across platforms. In line with open science principles, this review has been registered separately from a previously registered scoping review protocol, with explicit linkage maintained between the two registrations.