Skip to main content

Table 3 Performance of different parameter optimization methods using 3-fold cross-validation

From: Prediction of protein solvent accessibility using PSO-SVR with multiple sequence-derived features and weighted sliding window scheme

Method

PSAP300

PSAP2312

MAE (%)

PCC

MAE (%)

PCC

SVR

19.6

0.60

14.8

0.67

SVR-grid search1

17.3

0.67

14.7

0.69

PSO-SVR2

16.8

0.69

14.1

0.75

  1. 1kernel = Gaussian, C = 0.01, γ = 0.0025, ɛ = 0.05; 2kernel = Gaussian, C = 0.00762, γ = 0.00130, ɛ = 0.04129; best results are shown in bold.