You are viewing a single comment's thread from:
RE: Using Machine Learning to Fight Plagiarism
A decision to 'nuke' someone should never be algorithm based. Despite a fancy name 'machine learning', these algorithms are very primitive.
These tools can be used as an alerting system. For instance, if the bot finds a suspicious post it informs #steemitabuse chat. Humans can then validate the claim. If the author in question is a host of guest posts, then, he/she may be put on a 'whitelist'.