I have implemented a chat which filter bad words. I have a NoSQL database where I store all the users and chats, and basically my idea is to process all chats, filtering the swear words, and if found, pass this chat to the content moderators.
I have never done that, that is why I am asking, to get an idea of how this process is carried out in real life.
I have thought of creating a separate collection in the NoSQL database that only moderators have access to, and thus, put the bad chats as documents in it. I suppose this will be so, I don't know (maybe is not right to consume space in the app db). If yes, how do the professionals share these chats with the moderators? Do they give them direct access to the database or through an user interface?
The question may seem a bit absurd, or based on personal opinions, but it really is a question that I have. In other words, my intention here is to get an idea of how to achieve this process following good practices.
Thanks.