far-right-analysis icon indicating copy to clipboard operation
far-right-analysis copied to clipboard

Published 20 hours ago • verified publisher icon Data4Democracy

Hope speech classifier

Open gati opened this issue 7 years ago • 4 comments

"Hope speech" is counter-speech - messages posted in response to hate speech on a social platform. This task is to train a classification model that can identify hope speech.

One path might be, possibly using the classifier from #14, identify hate speech associated with a hashtag we can use the Assemble infrastructure to monitor, or the comments of a Facebook page we can use collect-social to monitor. Then look at the replies to that hate speech, and identify those that are in opposition to the original message. (Figuring out what's in opposition, or building a training set of counter-speech, is the tricky part).

Once you've built this, ping Jonathon (@gati) - next step will be to see which hope speech is effective.

gati avatar Mar 05 '17 22:03 gati

Is someone working on this ? If not, I'd like to take this up.

fx86 avatar Jun 13 '17 02:06 fx86

I don't think so, go for it @fx86! Please let me know if there's anything I can do to help!

gati avatar Jun 13 '17 13:06 gati

Have a few more questions. Can we do this on Slack ?

fx86 avatar Jun 14 '17 09:06 fx86

Sorry for the slow response! Pinged you back on Slack. Just give me a shout if I can help get you started. :)

On Wednesday, June 14, 2017, fx86 [email protected] wrote:

Have a few more questions. Can we do this on Slack ?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Data4Democracy/far-right-analysis/issues/15#issuecomment-308378609, or mute the thread https://github.com/notifications/unsubscribe-auth/AAhLiy2NGF8z3yzk0r39bCJ50lieLB8eks5sD6mBgaJpZM4MTkhk .

gati avatar Jun 17 '17 17:06 gati