Skip to Main Content
This paper introduces an approach of neural-fuzzy inference system (NFIS) with a genetic algorithm (GA) based on weighted data normalization (WDN) and its application in a medical decision support system. The WDN method optimizes the data normalization ranges for the input variables of the neural-fuzzy inference system and a genetic algorithm is used as part of the WDN method. A steepest descent algorithm (BP) is used for NFIS learning on the normalized data set. The derived weights have the meaning of feature importance and can be used for feature selection to decrease the number of input variables. The GAWDN-NFIS is illustrated on the case study: a real medical decision support problem of estimating the survival of haemodialysis patients. This approach can also be applied to other distance-based, prototype learning neural network or fuzzy inference models.