|
|
|
|
|
by PaulHoule
987 days ago
|
|
I am mostly interested in blocking for myself. What other people see doesn’t bother me. The one area where I’d like to put blocking into the community is with some kind of sentiment analysis, I think angry content is contagious and that it would be a lot of work to make an algorithmic feed more toxic than boosting + chronological feed + explore the most boosted. In fact I would start with making a predictive model of what succeeds in that environment and then filter away the non-toxic. If you could suppress angry toots in the platform, however I think it would lower the temperature level. I have looked for an open source sentiment analysis model for social media as opposed to product reviews and so far turned up empty as I’d like one for my own feed (though blocking f5t and t3s keywords works wonders). I have thought about making my own but collecting 5000 angry toots would probably kill me. |
|
That seems like an assertion more than something that can be proven.
Surely it is right to be angry about threats as anger motivates action to eliminate the threat.