Finally, based on the best feature combination, we developed one best predictive model set and applied it to predict the primary tumor sites.
Similarly, in the case of the GPCRs dataset, the NP kernel led to the best predictive model, with RMSEPext and R 0 ext 2 values of 0.66 p K i units and 0.72.
The best predictive model for death and sustained VT by multivariable analysis included RV mass-to-volume ratio ≥0.3 g/mL, LV EF z score <−2.0 (<55% in males and <54% in females), and history of atrial tachyarrhythmias (model c-index 0.832; R 0.246; table 4).
Test results demonstrate that the 2006 turbidity prediction model was selected as the best predictive model (RMSE = 5.787 and CC = 0.975) because of its ability to predict turbidity within the acceptable error range and 90%% required confidence interval (50NTU).
Taken together our results suggest that the ancestral state reconstruction procedure implemented in Bayesian framework following Organ et al. [ 55] might be more reliable in comparison to parsimony and ML methods, as it first identifies the best predictive model based on known data, and then uses such model to infer unknown ancestral states.
The 2006 turbidity prediction model was selected as the best predictive model because of its ability to predict turbidity within the acceptable error range and required confidence interval.
The best predictive model in terms of sensitivity and specificity was obtained with positivity at 3 out of 5 variables (sensitivity 61.4%, specificity 61.8%, and accuracy 62%).
Using all available variables the best predictive model of V included Hsd, PCveg and stand density (S) in the following formulation, V = 7.08 + 0.0690 H sd − 0.0423 P C veg + 0.893 1 − exp − 0.00691 S (6).
The inclusion of age, however, resulted in the best predictive model across all combinations of methods and models considered (Table 1).
We performed cross-validation to test the predictive ability of the different models and to find the best predictive model.
