This thesis work concerns the modelling of experimental data, regarding velocity fields obtained through Particle Image Velocimetry measurement method. Such data sets are correlated through an Adaptive Neuro Fuzzy Inference system, which is a non - deterministic type of model. This is achieved using a Fuzzy inference system structure, specifically a First - order Sugeno type. Modifiable parameters of the presented models are characterised by a hybrid machine Learning method, which iteratively takes a certain amount of data to train the initial raw model. Different types of models are created and analysed in terms of correlation and computational performances. This has the purpose of making considerations about the Neuro Fuzzy system itself.