Individual scatter partition-based fuzzy neural networks using particle swarm optimisation

This paper presents a new design of fuzzy neural networks (FNNs) based on individual scatter partition using particle swarm optimisation (PSO). The proposed FNNs are expressed by the scatter partition of input space generated by fuzzy c-means clustering algorithm. The partitioned local spaces indicate the fuzzy rules of the FNNs that have the individual regions in the different size. The consequence part of the rule is represented by polynomial functions. The back propagation algorithm is used to estimate the coefficients of the polynomial functions. The optimisation to find individual regions and parameters of learning is conducted by PSO. The performance of the proposed FNNs is demonstrated with the non-linear process.