A GA-Based Approach to ICA Feature Selection: An Efficient Method to Classify Microarray Datasets

Authors:
Kun-Hong Liu;Jun Zhang;Bo Li;Ji-Xiang Du
Affiliations:
School of Software, Xiamen University, Xiamen, China 361005;School of Electronic Science and Technology, Anhui University,;School of Computer Science of Technology, Wuhan University of Science and Techology, Wuhan, Hubei, P.R. China 430081;Department of Computer Science and Technology, Huaqiao University, Quanzhou, Fujian, P.R. China 362021
Venue:
ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part II
Year:
2009

Citing 4
Cited 0

Systematic benchmarking of microarray data classification: assessing the role of non-linearity and dimensionality reduction

Bioinformatics
Multiclass cancer classification and biomarker discovery using GA-based algorithms

Bioinformatics
Independent component analysis-based penalized discriminant method for tumor classification using gene expression data

Bioinformatics
Fast and robust fixed-point algorithms for independent component analysis

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Although many independent component analysis (ICA) based algorithms were proposed to tackle the classification problem of microarray data, a problem is usually ignored that which and how many independent components can be used to best describe the property of the microarray data. In this paper, we proposed a GA approach for IC feature selection to increase the classification accuracy of two different ICA based models: penalized independent component regression (P-ICR) and ICA based Support Vector Machine (SVM). The corresponding experimental results are listed to show that the IC selection method can further improve the classification accuracy of the ICA based algorithms.