Algorithm 3: Weighted random forest algorithm for dengue prediction

Input: Dengue Training Partition (P), Count of Trees (N), Features Subset—Random (FS) Output: Random Forest (RF) Tree with Dengue Prediction

For each i = 1 to N, do:

Apply bootstrap algorithm on training partition (P) such as ${\mathrm{P}}_{\mathrm{i}}=\mathrm{bootstrap} \left(\mathrm{P}\right).$

Apply the Decision Tree (DT), ${\mathrm{DT}}_{\mathrm{i}}=\mathrm{Random} \mathrm{Decision} \mathrm{Tree} \left(\mathrm{P}, \mathrm{FS}\right).$

Build the RF as $\mathrm{RF}=\mathrm{RF}\cup {\mathrm{DT}}_{\mathrm{i}}$.

End for

For each i = 1 to N, do:

Calculate the weight ($w_i^t$) of ith sample using Equation (1). $$w_i^t=\frac{1}{\mathrm{OB}}{\displaystyle \sum }_{\mathrm{j}ϵ\mathrm{OB}}\left|{\mathrm{X}}_{\mathrm{pred}}{ }_{\mathrm{i},\mathrm{j}}-{\mathrm{X}}_{\mathrm{actual}}{ }_{\mathrm{i},\mathrm{j}}\right|$$ (1)

End for

For each i = 1 to N, do:

$${\varnothing }_{\mathrm{i}}=\mathrm{f}\left(\mathrm{AUC}{\left({\mathrm{w}}^{\mathrm{t}}\right)}_{{\mathrm{IB}}_{\mathrm{i}}},\mathrm{AUC}{\left({\mathrm{w}}^{\mathrm{t}}\right)}_{{\mathrm{oB}}_{\mathrm{i}}}\right)$$ (2)

End for

For each i = 1 to, do:

Calculate the weight (${\mathrm{w}}_{\mathrm{i}}$) using Equation (3). $${\mathrm{w}}_{\mathrm{i}}=\frac{{\left(\mathrm{N}{\displaystyle \sum }{\mathrm{p}}_{\mathrm{i}}+1\right)}^{\mathrm{k}}}{{{\displaystyle \sum }}_{\mathrm{k}=1}^{\mathrm{NT}}{\left(\mathrm{N}-{\mathrm{p}}_{\mathrm{i}}+1\right)}^{ \mathrm{k} \prime }}$$ (3)

For each i = 1 to N, do:

Calculate the Final Prediction using Equation (4). $$\mathrm{X}\_{\mathrm{pred}}_{\mathrm{i}}=\frac{1}{\mathrm{NT}}{\displaystyle \sum }_{\mathrm{j}=1}^{\mathrm{NT}}\mathrm{X}\_{\mathrm{pred}}_{\mathrm{i},\mathrm{j}}\times {\mathrm{w}}_{\mathrm{j}}$$ (4)

End for

Return RF.