Evolutionary Bayesian Probabilistic Neural Networks

A well-known and widely used model for classification and prediction is the Probabilistic Neural Network (PNN). PNN’s performance is influenced by the kernels’ spread parameters so recently several approaches have been proposed to tackle this problem. The proposed approach is a combination of two well known methods applied to PNNs. First, it incorporates a Bayesian model for the estimation of PNN’s spread parameters and then an evolutionary optimization algorithm is used for the proper weighting of PNN’s outputs. The Particle Swarm Optimization (PSO) algorithm is used for a better estimation of PNN’s prior probabilities. The new model is called Evolutionary Bayesian Probabilistic Neural Network (EBPNN). Furthermore, a different kernel function, namely the Epanechnikov kernel, is used besides the typical Gaussian kernel. The above approach is applied to two biomedical applications with encouraging results and is compared with Feed-Forward Neural Networks.

[1]  Russell C. Eberhart,et al.  A new optimizer using particle swarm theory , 1995, MHS'95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science.

[2]  Donald Geman,et al.  Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images , 1984 .

[3]  Nicos G. Pavlidis,et al.  New Self-adaptive Probabilistic Neural Networks in Bioinformatic and Medical Tasks , 2006, Int. J. Artif. Intell. Tools.

[4]  Donald F. Specht,et al.  Probabilistic neural networks , 1990, Neural Networks.

[5]  E. Parzen.  On Estimation of a Probability Density Function and Mode , 1962 .

[6]  Michael N. Vrahatis,et al.  Recent approaches to global optimization problems through Particle Swarm Optimization , 2002, Natural Computing.

[7]  Carl G. Looney,et al.  A Fuzzy Classifier Network with Ellipsoidal Epanechnikovs , 2001 .

[8]  Michael N. Vrahatis,et al.  On the computation of all global minimizers through particle swarm optimization , 2004, IEEE Transactions on Evolutionary Computation.

[9]  Lutz Prechelt,et al.  A Set of Neural Network Benchmark Problems and Benchmarking Rules , 1994 .

[10]  David J. Hand,et al.  Kernel Discriminant Analysis , 1983 .

[11]  Catherine Blake,et al.  UCI Repository of machine learning databases , 1998 .

[12]  Ali S. Hadi,et al.  Finding Groups in Data: An Introduction to Chster Analysis , 1991 .

[13]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[14]  Chenn-Jung Huang.  A PERFORMANCE ANALYSIS OF CANCER CLASSIFICATION USING FEATURE EXTRACTION AND PROBABILISTIC NEURAL NETWORKS , 2002 .

[15]  Jian Guo,et al.  A Novel Method for Protein Subcellular Localization Based on Boosting and Probabilistic Neural Network , 2004, APBC.

[16]  Dimitris K. Tasoulis,et al.  Generalized locally recurrent probabilistic neural networks with application to text-independent speaker verification , 2007, Neurocomputing.

[17]  Todor Ganchev,et al.  Locally Recurrent Probabilistic Neural Networks with Application to Speaker Verification , 2004 .