Hierarchical Fuzzy Clustering in Conjunction with Particle Swarm Optimization to Efficiently Design RBF Neural Networks

摘要: A new method that combines hierarchical fuzzy clustering and particle swarm optimization is proposed to elaborate on an effective design of radial basis functions neural networks. As a first step, we pre-process the available data using ordinary partitions defined input-output space generate aggregate subspaces uniformly cover space. We, then, put in place reform aforementioned terms weighted version c-means algorithm. The network's kernel centers are elicited by projecting resulting clusters input widths connection weights estimated via implementation optimization. To this end, novelty our contribution relies way manipulate information order investigate relationships context considering optimizer as major parameter estimation platform. Finally, modelling capabilities effectiveness network demonstrated through several experiments 10-fold cross-validation analysis.