The Mudcat Café TM
Thread #166020 Message #3992365
Posted By: Mr Red
14-May-19 - 02:14 PM
Thread Name: BS: A solution to Fakebook
Subject: RE: BS: A solution to Fakebook
Well lookye here:
A team at IBM have devized an AI/Algorithm that changes offensive words and renders the text more polite. The success rate isn't quoted but cames to outperform other attempts. They only mention Twatter & Reddit as the test media.
OK it is censorship (make of that what you will) but if successful it would be a good cyber trap for what all** reasonable peeps would accept.
** OK the field is open for people to declare their reasonableness or otherwise. Not that they need an open invite, as they have already declared.