If this app works as well as it says it could be truly beneficial to a lot of people. I would love to understand more about how it was built and how it works.
@tese_omesan Hi Tese, thank you for the kind words! The users beta testing this have had a good experience so far, but obviously there is a lot left to do!
We use sentiment analysis to determine the context of offending content (as best as we can) to hopefully make sure we aren't over-filtering but still delivering our promise.
Also you can check out our repo here: github.com/GetSoothe/soothe
@nikodraca awesome! Thank you! I will be trying it out and sharing it with my artivist circles. ❤️ I also talked about on my anchor channel and turned it into a moment on here https://twitter.com/theblacktck/...
@nassaraf Hi Naomi! We're actually using sentimood.js (https://github.com/soops/sentimood) to perform the sentiment analysis.
I'm not entirely sure what the accuracy % is but in terms of UX it seems to work well enough to detect negative contexts the majority of the time. This is something we plan on further developing with more sophisticated tech.
Hinge Wrapped
Hinge Wrapped
MailKing by cloudHQ
Hinge Wrapped