Feature Selection Method Based on Partial Least Squares and Analysis of Traditional Chinese Medicine Data

Table 6

Comparative analysis of several parameter combinations for 6 datasets.

	R^2
WYHXB
s = 0.001, k = 13	0.7366
s = 0.005, k = 13	0.7366
s = 0.01, k = 13	0.7427
s = 0.05, k = 13	0.7403
s = 0.1, k = 13	0.7434
s = 0.11, k = 13	0.7434
s = 0.12, k = 13	0.7434
s = 0.13, k = 13	0.7418
s = 0.14, k = 13	0.7418
s = 0.15, k = 13	0.7418
s = 0.20, k = 13	0.7235
NYWZ
s = 0.001, k = 12	0.7361
s = 0.005, k = 12	0.7361
s = 0.01, k = 12	0.7435
s = 0.05, k = 12	0.7456
s = 0.1, k = 12	0.7689
s = 0.11, k = 12	0.7692
s = 0.12, k = 12	0.7692
s = 0.13, k = 12	0.7468
s = 0.14, k = 12	0.7468
s = 0.15, k = 12	0.7468
s = 0.20, k = 12	0.7344
DCQT
s = 0.001, k = 10	0.8760
s = 0.005, k = 10	0.8760
s = 0.01, k = 10	0.8760
s = 0.05, k = 10	0.8760
s = 0.1, k = 10	0.9546
s = 0.11, k = 10	0.9546
s = 0.12, k = 10	0.9546
s = 0.13, k = 10	0.9546
s = 0.14, k = 10	0.9546
s = 0.15, k = 10	0.9546
s = 0.20, k = 10	0.9546
CCrime
s = 0.001, k = 12	0.6524
s = 0.005, k = 12	0.6524
s = 0.01, k = 12	0.6722
s = 0.05, k = 12	0.6625
s = 0.1, k = 12	0.6708
s = 0.11, k = 12	0.6620
s = 0.12, k = 12	0.6678
s = 0.13, k = 12	0.6678
s = 0.14, k = 12	0.6678
s = 0.15, k = 12	0.6678
s = 0.20, k = 12	0.6678
BreastData
s = 0.001, k = 12	0.6419
s = 0.005, k = 12	0.6419
s = 0.01, k = 12	0.6419
s = 0.05, k = 12	0.7401
s = 0.1, k = 12	0.7436
s = 0.11, k = 12	0.7436
s = 0.12, k = 12	0.7436
s = 0.13, k = 12	0.7887
s = 0.14, k = 12	0.7887
s = 0.15, k = 12	0.7887
s = 0.20, k = 12	0.7213
RBuild
s = 0.001, k = 13	0.9407
s = 0.005, k = 13	0.9407
s = 0.01, k = 13	0.9589
s = 0.05, k = 13	0.9845
s = 0.1, k = 13	0.9593
s = 0.11, k = 13	0.9593
s = 0.12, k = 13	0.8803
s = 0.13, k = 13	0.8803
s = 0.14, k = 13	0.8803
s = 0.15, k = 13	0.9285
s = 0.20, k = 13	0.9285