Information visualization aimed at facilitating human perception is an important tool for the interpretation of experiments on the basis of complex multidimensional data characterizing the operational space of fusion devices. This work describes a method for visualizing the operational space on a two-dimensional map and applies it to the discrimination of type I and type III edge-localized modes (ELMs). The approach accounts for stochastic uncertainties that play an important role in fusion data sets, by modelling measurements with probability distributions in a metric space. The method is aimed at contributing to physical understanding of ELMs as well as their control. Furthermore, it is a general method that can be applied to the modelling of various other plasma phenomena as well.