Skip to main content

Table 5 Top 10 important descriptors selected by three blending methods

From: Comparison and improvement of the predictability and interpretability with ensemble learning models in QSPR applications

Uniform blendingLinear blendingAny blending
Selected descriptorsFeature importanceSelected descriptorsFeature importanceSelected descriptorsFeature importance
Gap0.1654Gap0.1330Gap0.1572
AP(xx)0.0470AP(xx)0.0359AP(xx)0.0411
F01[C-N]0.0216F01[C-N]0.0247Solvent0.0224
SpMax2_Bh(i)0.0176SpMax2_Bh(i)0.0174F01[C-N]0.0198
Solvent0.0169Solvent0.0141SpMax2_Bh(i)0.0170
P_VSA_MR_70.0162F02[C-N]0.0133P_VSA_ppp_L0.0147
P_VSA_ppp_L0.0151EP(xx)0.0122EP(xx)0.0115
EP(xx)0.0133F01[C-C]0.0111F02[C-N]0.0108
Chi1_EA(dm)0.0119SpMin5_Bh(m)0.0106Chi1_EA(dm)0.0105
F02[C-N]0.0109P_VSA_ppp_L0.0103Chi0_EA(dm)0.0092
  1. Italic values indicate the significance of important features which highly affect the fluorescence wavelengths