The Discourse moderation system has a weighting system that works in the way you are describing. The likelihood that a flagged post will be hidden is based on the reliability of the flagger’s previous flags. The score is determined by the percentage of times that a user’s flags have been agreed with by a staff member.
I don’t believe that the flag weighting system is documented on Meta. One of my goals for the coming week is to try to document it.
Edit: just adding a note that this topic does a good job of documenting how flags are weighted depending on a user’s prior flags: User Reputation and Flag Priorities.