Figure 2From: Principal component analysis-based unsupervised feature extraction applied to in silico drug discovery for posttraumatic stress disorder-mediated heart diseaseMatthews correlation coefficients and F measure for various FE methods applied to the simulated data set.t test, one vs one t test based FE; CRP, categorical regression based FE (using adjusted P-values); CRR, categorical regression based FE (using ranked P-values); BAHSIC, backward elimination using Hilbert-Schmidt norm of the cross-covariance operator; VBPCAFE, variational Bayes principal component analysis based unsupervised FE; CPCAFE, conventional principal component analysis based unsupervised FE.Back to article page