Table 2 The accuracy of Deep Belief Network, Support Vector Machines, and Neural Networks in terms of Mean Absolute Error (MAE) based on cross validation of training datasets with 16 features, the average per-target correlation, and loss on stage 1 and stage 2 of CASP11 datasets for all three difference techniques. P-value is calculated for the significance of DBN compared to other two methods
From: DeepQA: improving the estimation of single protein model quality with deep belief networks
| Â | MAE based on cross validation | Corr. on stage 1/significance of P-value | Loss on stage 1/significance of P-value | Corr. on stage 2/significance of P-value | Loss on stage 2/significance of P-value |
|---|---|---|---|---|---|
Deep Belief Network | 0.08 | 0.63/- | 0.09/- | 0.34/- | 0.06/- |
Support Vector Machine | 0.12 | 0.58/1.97E-01 | 0.10/6.17E-01 | 0.32/4.45E-04 | 0.07/7.41E-01 |
Neural Network | 0.08 | 0.51/9.74E-04 | 0.12/8.35E-02 | 0.25/1.05E-05 | 0.07/1.19E-01 |
Mean | 0.09 | 0.57/9.88E-02 | 0.10/3.50E-01 | 0.30/2.28E-04 | 0.07/4.30E-01 |