Feature Selection Using Hybrid Evaluation Approaches Based on Genetic Algorithms

For a given set of samples, a new model is proposed to reduce input feature space, which decreases the learning time of classifiers, but also, improves the prediction accuracy according to the chosen relevance criterion. This model is constructed by decision trees and genetic algorithms, which evalu...

Full description

Saved in:
Bibliographic Details
Published inElectronics, Robotics and Automotive Mechanics Conference (CERMA'06) Vol. 2; pp. 245 - 250
Main Authors Giraldo, T.L.F., Delgado, T.E., Riano, J.C., Castellanos, D.G.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2006
Subjects
Accuracy
Classification tree analysis
Decision trees
Feature extraction
Genetic algorithms
Nearest neighbor searches
Pathology
Performance analysis
Predictive models
Speech
Online AccessGet full text

Cover

More Information
Summary:For a given set of samples, a new model is proposed to reduce input feature space, which decreases the learning time of classifiers, but also, improves the prediction accuracy according to the chosen relevance criterion. This model is constructed by decision trees and genetic algorithms, which evaluates by means of k nearest neighbor rule for classification, allowing the evolution model parameters of used genetic algorithm. The training set corresponds to the extracted features from pathological (hypernasality) and non-pathological (normal) speech, acquired from 90 children, 45 examples per class. A comparative analysis between different approaches about feature selection is performed upon experimental results, showing the feasibility of this approach in such a cases involving pathologies recognition
ISBN:9780769525693
0769525695
DOI:10.1109/CERMA.2006.113