For the sake of simplicity, these values will not be repeated in Tables 8, 9, and 10, which present the communication performance when the default values of HWMP parameters are changed.
Preferred returns ensure that a fund manager only receives performance-based remuneration when the value of the fund exceeds the pre-agreed threshold (eg, 8%).
However, the IIC method has the similar performance as ID-FSD only when the value of α is large.
On the other hand, providing a dynamic spatial link between graphical elements of the configural display and the digital value improves performance when the variable of interest is dependent upon data relationships.
The approximation d̂SPR is very good for small values of dSPR, and we observed a lower performance when the "real" number of SPR moves increases.
As such, we also provide performance results when varying the value of T. Figure 10 shows how the throughput varies with respect to T, for the deterministic cases (Case D.1 to Case D.4).
The better performance of the algorithm comes when the values of FNR and FPR are smaller.
Although the values of λ vary a litter when the values of PPV, TPR and MCC reach their peak, DCS remains high performances when λ is in the range of 0.05 to 0.35.
We used random forest attribute importance measures in this step, which are based on the decrease of classifier performance when values of a variable in a bifurcating tree node are permuted randomly [ 80], implemented in the extended version of weka 3.5.1 [ 81, 82] (More details in Additional file 4).
Figure 1(b) showed that there is no significant (p-value = 0.469 using ANOVA) decrease of prediction performances, when the number of features is filtered down.
