Optimization without losing the accuracy and interpretability of rules is a major concern in rule based system. Fuzzy Inference system characterized by uncertainty tolerance is the best way to represent a knowledge based system. Optimization of rule based systems starts by incorporating selflearning ability to a fuzzy inference system. This can be achieved by neural networks, there by developing a neuro fuzzy inference system. This paper analyses different neuro fuzzy inference systems.The analysis has been performed in different types of datasets in terms of dimensionality and noises. Analysis results concludes that the neuro fuzzy model DENFIS (Dynamically Evolving Neuro Fuzzy Inference System) shows an improved performance when handling with high dimensional data. Simulation results on low dimensional data exhibits similar performance in ANFIS (Adaptive Neuro Fuzzy Inference System) and Denfis.
cited By 0
A. J., Radha D., and S., S., “Analysis of fuzzy rule optimization models”, International Journal of Engineering and Technology, vol. 7, pp. 1564-1570, 2015.