A comparative analysis across extremist Reddit communities

Am I missing or misinterpreting something?

This study's whole analysis, including comparison of toxicity in misandrist and misogynist subreddits, is based on an output of a model fine-tuned on hate speech, annotated in the following manner:

> Some identities were considered out-of-scope for Hate, including men, white people, and heterosexuals

Reddit is infested with bots and trolls, I avoid it as much as I can.

Frustrating that Google promoted it so much.