A hybrid GA & back propagation approach for gene selection and classification of microarray data

Authors:
Omid Khayat;Hamid Reza Shahdoosti;Ahmad Jaberi Motlagh
Affiliations:
Biomedical eng., Computer eng., Amirkabir University;Biomedical eng., Computer eng., Amirkabir University;Biomedical eng., Computer eng., Amirkabir University
Venue:
AIKED'08 Proceedings of the 7th WSEAS International Conference on Artificial intelligence, knowledge engineering and data bases
Year:
2008

Citing 8
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Gene Selection for Cancer Classification using Support Vector Machines

Machine Learning
Estimating the Generalization Performance of an SVM Efficiently

ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
An introduction to variable and feature selection

The Journal of Machine Learning Research
A novel feature selection method to improve classification of gene expression data

APBC '04 Proceedings of the second conference on Asia-Pacific bioinformatics - Volume 29
HykGene: a hybrid approach for selecting marker genes for phenotype classification using microarray gene expression data

Bioinformatics
Comparison of Multiobjective Evolutionary Algorithms: Empirical Results

Evolutionary Computation
Enhancing genetic feature selection through restricted search and Walsh analysis

IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews

Quantified Score

Hi-index	0.00

Visualization

Abstract

We propose a Genetic Algorithm (GA) approach combined with Neural Network (MultiLayer Perceptron) with Back Propagation algorithm (BP) for the classification of high dimensional Microarray data. This approach is associated to a fuzzy logic based pre-filtering technique. The GA is used to evolve gene subsets whose fitness is evaluated by a NN classifier. Using archive records of "good" gene subsets, a frequency based technique is introduced to identify the most informative genes. Our approach is assessed on two well-known cancer datasets and shows competitive results with six existing methods.