In this paper one aspect of the plasma evolution control in tokamak (nuclear fusion) reactors is assessed, namely, the identification part of the controller. A fuzzy inference system (FIS) for plasma shape recognition applications is firstly presented. The model is directly extracted from a data set of examples of the problem in the absence of learning procedures. The most relevant advantages of the FIS are: 1) the solution of the problem can be expressed in terms of very simple as well as explainable rules, and 2) a very limited number of inputs is required to obtain a sufficient estimation accuracy. The first objective overcomes one of the most limitations of Neural Network (NN) models. The second one has a strong impact on the throughput time in real time applications. The resulting model can be tuned by varying the parameters of the membership functions (centres and variances of the Gaussian functions) in order to best fit the data set distribution. In this case, we shall have a neuro-fuzzy model, which will be more accurate with respect to the naive fuzzy model. The qualitative analysis of the data set carried out by using the fuzzy logic approach can also capture relevant insight on some difficult aspect of the problem, like its basic ill-posedness and the detection of category transition. The results presented in this paper regards a benchmark database of simulated plasma equilibria in the ASDEX-Upgrade machine. The main conclusion is that a FIS is by itself an efficient tool for real time analysis of magnetic data in tokamak reactors and that the neuro-fuzzy framework can yield models competitive with conventional statistical-based systems.