I'm drawing a Finite State Machine (which I'll have to implement later on) which is semi-automatic, i.e. some transitions require user input, some are decided on internal value.
The problem is that some transitions are decision trees with several possible transitions. How should I draw such cases? I am considering drawing something like submachine, representing the decision tree, but maybe there is some another more standard approach?
Decision trees apply only to user input. Decision tree is a series of yes/no questions and every answer might branch-off into another question or transition. There are no "empty" branches - transition will happen, though some transitions are duplicated, i.e. several decision paths may lead to the same transition.
The decision tree is at most 5 levels deep, therefore listing every combination of criteria on a transition is an overkill.