Establish a universally accepted definition of online misogyny

Establish a universally accepted definition of online misogyny that can be consistently applied across studies to guide manual annotation and automated detection of misogynistic language on social media.

Background

For classifier training and evaluation, the authors had to operationalize misogyny for manual coding in Brazilian Portuguese. They note that existing literature lacks consensus on a unified definition of online misogyny, complicating annotation and model development.

A common, widely accepted definition would improve comparability across studies, enable clearer coding instructions, and support the development of more robust machine-learning classifiers.