riot should fix this system because its completely unfair because some random people where screaming and me and reported me and im the one that got banned riot fix ur shit man because your going to lose a lot of players
ModThe Djinn6 votesDo you have your chat logs? It's extremely rare for the system to improperly ban someone and, while it can happen, it's uncommon enough that seeing the logs is important before we can recommend a course of action.
Your match history doesn't seem to throw any intentionally feeding flags, which means this is likely a behavioral punishment.
zPOOPz1 votesI don't mean to nitpick, but isn't intentionally feeding is a behavior issue as well?
ModThe Djinn1 votesI don't mean to nitpick, but isn't intentionally feeding is a behavior issue as well?
Well, yes, but it's not handled by the same behavioral system, to my knowledge. It's spotted by something other than the IFS.
Silent Gravity1 votesThe IFS is the system as a whole. It is made up of many different modules that each look for different things. It's just easier to call the chat modules "IFS" than it is to refer to each by their individual, unknown, names.
[{quoted}](name=Riot Tantram,realm=NA,application-id=ZGEFLEUQ,discussion-id=3zQ7hXi0,comment-id=00040007000000000000000100000000000000020000,timestamp=2017-11-17T07:50:43.899+0000)
You know about all of them.
The IFS (Instant Feedback System) system calls out to multiple services for evaluation. I should probably write a tech blog on this. There are different services that evaluate verbal harassment, zero tolerance, intentional feeding, afk'ing, etc.
Let's say that tomorrow we needed a new service to evaluate ... morse code. We would write a new service module and register it with the IFS. We simplify the discussion and just usually say IFS. But the reality is there is a very complex group of systems that support player behavior.
[{quoted}](name=Riot Tantram,realm=NA,application-id=ZGEFLEUQ,discussion-id=ssrwyraZ,comment-id=0003000100000000,timestamp=2017-12-12T23:40:16.406+0000)
There are numerous detection modules in the IFS, one of which is responsible for detecting a small number of zero-tolerance things such as hate speech. This is the most naive of the modules, since it only looks for things which should never have a place in game. Think of the hate speech module as a first pass detection to easily detect obvious things.
ModThe Djinn2 votes[{quoted}](name=Silent Gravity,realm=NA,application-id=ZGEFLEUQ,discussion-id=MAx3EXUU,comment-id=0001000000000000,timestamp=2018-03-10T23:33:56.913+0000)
The IFS is the system as a whole. It is made up of many different modules that each look for different things. It's just easier to call the chat modules "IFS" than it is to refer to each by their individual, unknown, names.
You're totally right. Got my terminology confused there. Thanks!
Eye of Harmony1 votesExtremely rare ? You must be joking lol... Its mathematically impossible to make a good system like this...
Even with the highest skills and programmers you would just get a system that will "usually" ban the good persons...
Eye of Harmony1 votesOh and i forgot to add, the bias come from the fact that so many people are toxic, those who are getting punished think its their fault, even if they are usually not the most toxic players...
edit: Being toxic will get you banned one day eventually but... Traits that will get u punished (edit: while not really flaming) are things like: Writing too much... Playing heroes that make other people flaming you... Talk back when flamed... Playing more that usual
Obviously that is a reasoning that is right for ranked game... For normal games maybe it works 'not that bad', since most of the problem that make the system not working would be less likely to appear
zPOOPz2 votes"usually" based on what metric and which standards?
Eye of Harmony1 votesYou want a demonstration ?
zPOOPz4 votessure!
Eye of Harmony1 votesIts not easy but i tried
Whatever i guess its better to just think about the problem: What is supposed to do a system that ban people like this ?
First question would be, on what can a system like this really work... Easy data you can gather are chat logs and game stats...
Problem is you cannot really use game stats, like the number of death, the picks... To really determine if someone is bannable...
Like, feeding or trollpick isn't a bad behaviour... Maybe you just don't play good today. Maybe you bored and want to play another champion.
Only intentional feeding is bannable. But its not easy to understand if someone is intentional feeding from the data of the game.
You can ban people on keywords. But here again the problem is to evaluate in which context it has been said.
When you have a community you can simple ban insults because you know that you do not want to see those words in ANY context
But imagine if someone say: "you are a @#$!", and you answer "why do you call me a @#$! ?"
A perfect system would have to understand that the second sentence is a question and not an affirmation...
Another exemple: the word "retarded". It can be an insult, but it can be a way to say that something is extraordinaly like "omg its so retarded !"
So if u understand that unless they got the best AI in the world, they have a simpler system...
A system that will count the number of bad words, the number of reports, try to value if the one sending a report is toxic too or is suppoed to be a good guy... And try to correlate those data...
So it will be very hard for them to understand the meaning of the grammar in context.
They will implement things like: If the message is just "retards" then its an insult, but if there is another word in the sentence then it is potentially NOT an insult. That is why saying "kys" is insta ban but saying "i think you should probably kill yourself" is not insta ban In this exemple if i say: "i think" "the ennemy team" "is" "retarded" i will count as extremely toxic, while i would simple be not polite as slightly stupid but not really "bannable"
And about some the report system...
So what this system do is basically the same as the downvote system... See my post are downvoted because i "seem" arrogant.
So with a system like this i would get banned because people don't like my point of view
This is more or less what finally happens with a report system like this ^^
All those problem are simply extremely difficult to solve !
Another exemple:
In ranked so many people flame...
But you get reported usually when game is lost. It means if you carry all game you get less reported ! Even if you are a flamer...
Another exemple: They all flamed but one get punished... That person will obviously think its normal he gets punished ! And yes it is... But other don't get punished and also think its normal they didn't get punished. (its called bias)
also i found an article that is maybe able to bring some insight " https://en.wikipedia.org/wiki/Philosophy_of_artificial_intelligence " i didn't read it though...
ModThe Djinn6 votes[{quoted}](name=Eye of Harmony,realm=EUW,application-id=ZGEFLEUQ,discussion-id=MAx3EXUU,comment-id=00010001,timestamp=2018-03-10T18:37:31.767+0000)Extremely rare ? You must be joking lol...
Given my experience on this boards with "undeserved" bans and the chat logs that follow, no. In my experience at it is very uncommon for a ban to go out that is not supported by problematic behavior.
What behavior the individual punished party feels is problematic may vary, but the system seems very consistent.
Even with the highest skills and programmers you would just get a system that will "usually" ban the good persons...
This seems contrary to what I know of current AI systems, programming techniques, and Riot's particular system
ModThe Djinn1 votes--Accidental Post. Oops!--
Eye of Harmony1 votesWell i'm sorry to have talked to you in such a sarcastic way its just i got a little frustrated from some other posts so... My excuses
Also i put an explanation of how i see that in answer to zPOOPz
Terozu1 votes.07% chance of a false punishment. Bans make up less than a third of those. And so out of about maybe 70 something bans a month, thats... 2, maybe 3 false bans monthly. And thats not counting that one of those should be a 14 day ban instead of a permanent.
Timethief492 votesYou dont get banned without doing anything, reports only trigger a review of the case. Feel free to post your chatlog and inform us about your previous punishments if you want anything here.
KingNothîng1 votesPerma ban for a one time offense. Ok. Curious how many other times you've been banned.
exitforever1 votesLogs? Also, were you 14 day banned before?
Kei1431 votesWhat does it say when you log in?
Xidphel1 votesI got Perma Banned** for no reason**
because i was feeding
CoopA8261 votesDo you think OP knows what "because" means?
Leaf on Bush1 votesLet me guess, they screamed at you and you screamed back. You got banned for being toxic, not for feeding.
