Hi everybody that is going to read this :)

I ended up here trying to report a guy and found out all the beautiful logs you guys gather on all the players, and it had me wondering... You have a big amount of really good data in your hands, do you have some kind of AI/Machine Learning algorithm going through all this data? I think it would be cool (and effective) to have some kind of automated tool that gives some score to potential cheaters, to match with people's reports.

Obviously if someone "in charge" likes the idea, I'd love to contribute to the awesome community you've built so far :) 

Cheers 

[EDIT]: I wish I could edit the title to "[...]CHEATER detection", but oh well... :)

The idea has been thrown around before but the general consensus is that there are too many edge cases that would need to be considered in the algorithm which would add to the complexity of the algorithm and make the algorithm less reliable to the point that each case would need to be manually audited, therefore at this point in time it would provide very little if any real benefit to the ban and appeal processes. 

Makes complete sense. For this very problem, these algorithms usually work as "decision support", sort of like a Green/Red light, or a "score" aiding someone (i.e. a moderator) in their decisions. As, for example, in a bank when someone submits a "loan request", together with the AI scoring him as a potential "bad payer", there's always a human supervising the algorithm's decision.

Basically, it is not full automation, it's a way of limiting the effort on very clear cases, focusing it on edge cases, which would still need a human brain so to speak.

I hope this makes sense to you since I don't actually know how the actual process works internally. 

Let me know if there's any reason to discuss it further, or if it is something written in stone.

I was unjustly banned for no reason