Model order selection of a fuzzy logic system

Fuzzy logic systems can be beneficial for physiological and medical applications. One issue that arises with their use is how to select the appropriate number of parameters for adequate system representation. This paper presents an empirical technique to select the number of modeling parameters for a Neuro-Fuzzy Inference System (NFIS) applied toward modeling of heart rate variability. The technique is simple yet effective. Further work is needed to develop a model order selection technique that can be applied toward general fuzzy logic systems in a more theoretical context