Title :
Frankenstein PSO applied to neural network weights and architectures
Author :
De Lima, Natália Flora ; Ludermir, Teresa Bernarda
Author_Institution :
Centre of Inf., Fed. Univ. of Pernambuco, Recife, Brazil
Abstract :
In this paper we present the FPSOA-FPSOP a modified version of Frankstein Particle Swarm Optimization (FPSO), which is used to adjust the weights and architectures of a feed forward neural network. To evaluate the algorithm we used benchmark classification problems in medical care area. The results were compared with other algorithms which use the same methodology to find out the weights and architectures.
Keywords :
feedforward neural nets; health care; neural net architecture; particle swarm optimisation; pattern classification; FPSOA-FPSOP; Frankenstein PSO; Frankstein particle swarm optimization; classification problems; feedforward neural network; medical care; neural network architecture; neural network weights; Artificial neural networks; Classification algorithms; Heart; Nickel; Optimization; Particle swarm optimization; Topology; Classification Problems; Neural Networks; Particle Swarm Optimization;
Conference_Titel :
Evolutionary Computation (CEC), 2011 IEEE Congress on
Conference_Location :
New Orleans, LA
Print_ISBN :
978-1-4244-7834-7
DOI :
10.1109/CEC.2011.5949921
