Hey @remotehorst23
The main work of the AI is to "process" all information of a post on the blockchain.
The reputation of the hive is deliberately ignored.
Processing is done;
- What is the syntax?
- How are the posts structured? Did the author make any effort?
- Are there the same or similar posts on the account?
- Are the words known and common or rather unknown and unique?
A normal bot will fail, because the calculated intelligence is almost 0.
If it comes to this, the AI goes through all accounts that are related to this account.
With all information it is possible to calculate at the end whether the next mail will be spam or not.
Unfortunately I can't go very deep into the topic, because this knowledge also benefits the "evil" spammers and that's not the point of this thing.
Also a lot of information are trade secrets of the CommentCoin and will be partially released later. So some more patience ;)
So I could see with a database comparison that with 64 evaluated posts you have a 54% probability that your next post will not be spam. So you would have received a CC on your next post.
Of course you can trick this now, if you know it.
But I can also see that your posts would have gotten a 55%, but your relations to bots (@reverse-engineer, @dein-problem) downgrade you a bit.
So it's a kind of own reputation through a computer that sees what's happening in real time.
There's too much entanglement.
It's not an accurate way to describe "just the Al".
Thanks for your answer!
OMG I'm damned! 😱
I didn't know that 46% of my posts are considered spam! I always thought I only made one or two spam posts (when I started with hive / steem and didn't know how things work here). For me, it's obvious that this number is very wrong. It should be something like 96%.
These Bots have a relation with me and not the other way around! If a bot downvotes you regularly and you don't give a shit about it, send him one hive to show him a middle finger with a smile. That's what I did and now this makes me related to him? Ok, it only made 1% minus for me.
But do you know that I have ZERO control over what kind of Bot interacts with me? And did you know that when a bot interacts with a person and the person is not happy with the outcome, the person is likely to interact with the bot to try to prevent that somehow? Your AI needs to understand the difference in that. It won't be an easy task...
Ok, somehow I followed @reverse-engineer but how can you say he is a bot? And how can you say that bots are all bad when you use a bot to identify the quality of posts? That's a bit crazy, isn't it? It's like the emperor saying that all emperors in the world are bad and need to be captured.
Can your AI tell me, what kind of relationship I have with @dein-problem (except getting downvoted by him)? This would be great because I really want to know!
If you want to, we can talk in Discord to find a solution. I really want to help to make your AI better. And there we can talk german. I guess it's easier for us both.
Discord: Remotehorst23#6069