GA-based feature subset selection: Application to Arabic speaker recognition system

Feature Selection is an important task which can affect the performance of pattern classification and recognition. In this paper, we present a feature selection algorithm based on genetic algorithm optimization. The algorithm adopts classifier performance and the number of the selected features as h...

Full description

Saved in:
Bibliographic Details
Published in2011 11th International Conference on Hybrid Intelligent Systems (HIS) pp. 383 - 387
Main Authors Harrag, A., Saigaa, D., Boukharouba, K., Drif, M., Bouchelaghem, A.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2011
Subjects
Biological cells
Classification algorithms
Error analysis
Feature extraction
feature selection
genetic algorithm
Genetic algorithms
Speaker recognition
Speech
Online AccessGet full text

Cover

More Information
Summary:Feature Selection is an important task which can affect the performance of pattern classification and recognition. In this paper, we present a feature selection algorithm based on genetic algorithm optimization. The algorithm adopts classifier performance and the number of the selected features as heuristic information, and selects the optimal feature subset in terms of feature set size and classification performance. Experimental results on various speakers show that our algorithm can obtain better classification accuracy with a smaller feature set which is crucial for real time application and low resources devices.
ISBN:1457721511
9781457721519
DOI:10.1109/HIS.2011.6122136