Abstract:Because of high dimensionality, correlation among covariates, and noise contained in data, dimension reduction (DR) techniques are often employed to the application of machine learning algorithms. Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and their kernel variants (KPCA, KLDA) are among the most popular DR methods. Recently, Supervised Kernel Principal Component Analysis (SKPCA) has been shown as another successful alternative. In this paper, brief reviews of these popular techniques are presented first. We then conduct a comparative performance study based on three simulated datasets, after which the performance of the techniques are evaluated through application to a pattern recognition problem in face image analysis. The gender classification problem is considered on MORPH-II and FG-NET, two popular longitudinal face aging databases. Several feature extraction methods are used, including biologically-inspired features (BIF), local binary patterns (LBP), histogram of oriented gradients (HOG), and the Active Appearance Model (AAM). After applications of DR methods, a linear support vector machine (SVM) is deployed with gender classification accuracy rates exceeding 95% on MORPH-II, competitive with benchmark results. A parallel computational approach is also proposed, attaining faster processing speeds and similar recognition rates on MORPH-II. Our computational approach can be applied to practical gender classification systems and generalized to other face analysis tasks, such as race classification and age prediction.

Comparative Research on Methods of Dimensionality Reduction in High-Dimension Medical Data

A Brief Survey of Dimension Reduction

Supervised dimensionality reduction for big data

A Comparative Study of Dimensionality Reduction Methods for Large-Scale Image Retrieval.

CARE: Finding Local Linear Correlations in High Dimensional Data

Maximally Correlated Principal Component Analysis Based on Deep Parameterization Learning.

Comparison among dimensionality reduction techniques based on Random Projection for cancer classification.

Research and Comparison of Data Dimensionality Reduction Algorithms

L-PCA-based dimensionality reduction algorithm for high dimension images

A Review of Principal Component Analysis Algorithm for Dimensionality Reduction

Supervised Linear Dimension-Reduction Methods: Review, Extensions, and Comparisons

Comparative Study on Dimensionality Reduction in Large-Scale Image Retrieval

Sparse Kernel Entropy Component Analysis for Dimensionality Reduction of Biomedical Data

PAM: An Efficient Hybrid Dimension Reduction Algorithm for High-Dimensional Bayesian Network

A Comparison Study on Nonlinear Dimension Reduction Methods with Kernel Variations: Visualization, Optimization and Classification.

Correlated Components Analysis - Extracting Reliable Dimensions in Multivariate Data

Finding High-Order Correlations In High-Dimensional Biological Data

Semi-supervised Enhanced Discriminative Local Constraint Preserving Projection for Dimensionality Reduction of Medical Hyperspectral Images

Feature Selection Method Based on Mutual Information and Class Separability for Dimension Reduction in Multidimensional Time Series for Clinical Data

Research on Dimensional Reduction of Sparse Matrix Data Based on Information Entropy

Nonlinear Dimensionality Reduction for Discriminative Analytics of Multiple Datasets