Grey Relational Analysis
A quantitative analysis method for evaluating factors and supporting decision-making algorithms by measuring relational degrees between data sequences
Addressing Gene Expression Profile Analysis Problems from a Factor Analysis Perspective
This study approaches gene expression profile analysis through factor analysis methodology. To resolve stability issues in conventional Independent Component Analysis (ICA), we propose a DNA microarray data integrated classifier based on selective ICA. The implementation involves analyzing reconstruction errors in gene expression levels, selecting ICs with minimal reconstruction errors for sample reconstruction, then training multiple Support Vector Machine (SVM) base classifiers concurrently using the reconstructed samples. Final classification is achieved through majority voting among high-accuracy base classifiers. Experimental validation across three benchmark datasets confirms the method's effectiveness.
Factor Analysis and Principal Component Analysis
Factor Analysis is a statistical technique for extracting common factors from variable groups, while Principal Component Analysis is a multivariate statistical method that reduces multiple variables to a few composite indicators. From a mathematical perspective, PCA serves as a dimensionality reduction technique using orthogonal transformations to convert correlated variables into linearly uncorrelated principal components.