PoodankMcGee
Crohn's/ostomycel
★
- Joined
- May 1, 2018
- Posts
- 4,052
Came across another, more interesting glownigger study than the one I showed you yesterday. This one came out last August and was sponsored by the Swedish government's Defense Research Agency. They used Google's machine learning algortithm, BERT, to analyze MILLIONS of posts from this forum, looksmaxx, and lookism.net
To get right to the juicy bits, they derived a "toxicity level" for each forum, using Reddit as a control to represent the "normal internet" They then categorized the posts on each forum based on the target of the toxicity.
Lots of interesting findings to pick apart here. Even though this forum is apparently the most toxic in absolute terms, lookism and looksmaxx users attack each other more frequently and exhibit more ethnicity-related toxicity.
The study also analyzed user lifespan, the results validating greycel discrimination and the general suspicion of new users.
They go into some discussion speculating the reason for these differences, but I'd rather hear your thoughts. I will leave you with their semi-based conclusion that trying to shut down our forums is ineffective and incompatible with freedom of speech: straight from the glowniggers' mouth
Source: https://link.springer.com/article/10.1007/s43545-021-00220-8
Pelzer, B., Kaati, L., Cohen, K., & Fernquist, J. (2021). Toxic language in online incel communities. SN Social Sciences, 1(8), 1-22.
To get right to the juicy bits, they derived a "toxicity level" for each forum, using Reddit as a control to represent the "normal internet" They then categorized the posts on each forum based on the target of the toxicity.
Lots of interesting findings to pick apart here. Even though this forum is apparently the most toxic in absolute terms, lookism and looksmaxx users attack each other more frequently and exhibit more ethnicity-related toxicity.
The study also analyzed user lifespan, the results validating greycel discrimination and the general suspicion of new users.
Keep in mind their data only went through January 2020, and the events since then may have altered things. They also made this graph of user lifespans that is honestly brutal for veterancelsOn incels.co... User retention is generally low, as approximately half of the members stayed active for no more than one month, and indeed 70% of all members who ever posted are no longer active. The forum had attracted some 400 new members in the month before our retrieval. Around the one-year mark of activity the inactive users become a minority in their respective cohort, indicating that users who stay longer tend to become permanent members
They go into some discussion speculating the reason for these differences, but I'd rather hear your thoughts. I will leave you with their semi-based conclusion that trying to shut down our forums is ineffective and incompatible with freedom of speech: straight from the glowniggers' mouth
One question arises after studying the toxic environment on incel forums: How do we relate to the incel forums’ hateful messages and their links to violent acts? It is clear that it does not help to shut down the forums. Incel ideology has mostly been driven away from large social media, so moderation is effective in general, but the current dedicated incel forums are safe spaces that explicitly allow and encourage such toxic messages. They are maintained by technologically competent individuals who are sympathetic to the incel subculture, and if shut down these forums will just appear somewhere else under a new name. It would be virtually impossible to eliminate the spreading of incel ideology without endangering our open society. We must adapt to a reality where such toxic online environments continue to operate, and we need to focus on ways to reduce the harm they cause... Obtaining a better understanding of the sociopsychological mechanisms behind incel activity is necessary to be able to present alternative narratives and encourage deliberation rather than affective polarization.
Source: https://link.springer.com/article/10.1007/s43545-021-00220-8
Pelzer, B., Kaati, L., Cohen, K., & Fernquist, J. (2021). Toxic language in online incel communities. SN Social Sciences, 1(8), 1-22.