Resource-allocating probabilistic neuro-fuzzy network

In this paper, an architecture of a resourceallocating learning probabilistic neural network is considered. Construction and learning algorithms are proposed. The advantages of this network lie in the possibility of classification of data with substantially overlapping clusters. The construction algorithm significantly reduces the size of the network and tuning of the activation function parameters improves the accuracy of classification. Simulation results confirm the efficiency of the proposed approach in the data classification problems.

• Type: Conference
• Year: 2003
• Where: EUSFLAT
• Authors: Yevgeniy Bodyanskiy, Yevgen Gorshkov, Vitaliy Kolodyazhniy
• Keywords: Activation Function Parameters, EUSFLAT 2003, EUSFLAT 2007, Network Lie, Probabilistic Neural Network