Feature Selection and Neural Network for analysis of microstructural changes in magnetic materials

Authors:
J. A. Pérez-Benitez;L. R. Padovese
Affiliations:
Departamento de Engenharia Mecínica, Escola Politécnica, Universidade de São Paulo, Av. Prof. Mello Moraes, 2231, 05508-900 São Paulo, Brazil;Departamento de Engenharia Mecínica, Escola Politécnica, Universidade de São Paulo, Av. Prof. Mello Moraes, 2231, 05508-900 São Paulo, Brazil
Venue:
Expert Systems with Applications: An International Journal
Year:
2011

Citing 3
Cited 2

Adaptive floating search methods in feature selection

Pattern Recognition Letters - Special issue on pattern recognition in practice VI
Enhanced feature selection models using gradient-based and point injection techniques

Neurocomputing
Probabilistic neural networks and the polynomial Adaline as complementary techniques for classification

IEEE Transactions on Neural Networks

Fast feature selection aimed at high-dimensional data via hybrid-sequential-ranked searches

Expert Systems with Applications: An International Journal
A system for classification of time-series data from industrial non-destructive device

Engineering Applications of Artificial Intelligence

Quantified Score

Hi-index	12.05

Visualization

Abstract

This work presents a Feature Selection Algorithm for optimization of a Probabilistic Neural Network. The aim of this Probabilistic Neural Network is classifying signals obtained from magnetic material samples where two micro-structural parameters were changed simultaneously. The combination of the Feature Selection Algorithm with the Probabilistic Neural Network shows both, classification rate outcomes and classification speed, higher than those coming from traditional Probabilistic Neural Networks.