Fuzzy clustering-based polynomial radial basis function neural networks (p-RBF NNs) classifier designed with particle swarm optimization

  • Authors:

  • Wook-Dong Kim;Sung-Kwun Oh;Hyun-Ki Kim

  • Affiliations:

  • Department of Electrical Engineering, The University of Suwon, Hwaseong-si, Gyeonggi-do, South Korea;Department of Electrical Engineering, The University of Suwon, Hwaseong-si, Gyeonggi-do, South Korea;Department of Electrical Engineering, The University of Suwon, Hwaseong-si, Gyeonggi-do, South Korea

  • Venue:
  • ISNN'11 Proceedings of the 8th international conference on Advances in neural networks - Volume Part I
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we introduce polynomial-based Radial Basis Function Neural Networks (p-RBF NNs) classifier based on Fuzzy C-Means (FCM) clustering method. The parameters (fuzzification coefficient of FCM and polynomial type of models) are optimized by means of Particle Swarm Optimization (PSO). The fitness of hidden layer is expressed in term of partition matrix resulting from fuzzy clustering in this case being FCM. As weights between hidden layer and output layer, four types of polynomials are considered. The performance of proposed model is affected by some parameters such as the fuzzification coefficient of the fuzzy clustering (FCM) and the type of polynomial between hidden layer and output layer. The parameter coefficient of polynomial (weight) is obtained by using Weighted Least Square Estimation (WLSE) to improved performance and interprebility of local models. The proposed classifier is applied to a synthetic and machine learning dataset and its results are compared with those reported in the previous studies.