ICA and PCA Integrated Feature Extraction for Classification

Accurate feature extraction plays a vital role in the fields of machine learning, pattern recognition and image processing. Feature extraction methods based on principal component analysis (PCA), independent component analysis (ICA), and linear discriminant analysis (LDA) are capable of improving the performances of classifiers. In this paper, we propose two features extraction approaches, which integrate with the extracted features of PCA and ICA through some statistical criterion. The performances of the proposed feature extraction approaches are evaluated on simulated data and three public data sets by using cross-validation accuracy of different classifiers that found in statistics and machine learning literature. Our experiment result shows that integrated with ICA and PCA feature is more effective than others in classification analysis.