How does the IFS learn?
**First question: **Does the IFS ever review a game without reports?
**Second question: **How does the IFS determine what is punishable?
Third question: How does the IFS determine what is not punishable?
15 Comments
Given my knowledge of IFS, here's my take on the questions you asked.
**First question: **Does the IFS ever review a game without reports?
In some cases, yes. Intentional feeding or extreme racism in chat don't need a report to trigger a review.
**Second question: **How does the IFS determine what is punishable?
Based on data from previous games/reports, input from players, and parameters set by the designers.
Third question: How does the IFS determine what is not punishable?
Based on input from players and parameters set by the designers.
It's not so much "if X exists, then punish", it's more "if negative behavior exists, then begin review" and what happens during the review process determines if a punishment is handed out. You will never be able to find a strict, comprehensive list of things you can/can't do. For one, it would be way too long, and for two, you can always find ways around the existing list to proliferate negative behavior, so keeping a list would be too much work for little benefit.
its reviews games that are reported and gets hot words that are commonly used by toxic players. Some words are considered never to be toxic like the or i. lets use kys for example. say over the next week kys is used in 300 games. Of those 300 games the player who used it is reported 270 times. The system can now add it as a toxic word.
It doesn't. In reality its just a system that scans for certain keywords.