Algorithm |
\(R^{2}_{CV}\)
| RMSECV
|
\(R^{2}_{0\ test}\)
| RMSEtest
|
---|
A |
GBM | 0.59 | 0.77 | 0.60 | 0.76 |
RF | 0.60 | 0.78 | 0.61 | 0.79 |
SVM | 0.61 | 0.75 | 0.60 | 0.76 |
B |
Greedy ensemble | – | 0.73 | 0.63 | 0.73 |
Linear stacking | 0.63 | 0.73 | 0.63 | 0.73 |
EN stacking | 0.63 | 0.72 | 0.62 | 0.72 |
SVM linear stacking | 0.63 | 0.73 | 0.62 | 0.73 |
SVM radial stacking | 0.63 | 0.73 | 0.63 | 0.73 |
RF stacking | 0.61 | 0.76 | 0.58 | 0.77 |
- Combining single models trained with different algorithms in model ensembles allows to increase model predictive ability. We obtained the highest \(R^{2}_{0\ test}\) and RMSEtest values namely, 0.63 and 0.73 pIC50 unit respectively, with the greedy ensemble, and with the following model stacking techniques: (1) linear, and (2) SVM radial.
-
EN Elastic Net, GBM Gradient Boosting Machine, RF Random Forest, RMSE root mean square error in prediction, SVM Support Vector Machines.