Algorithm 1: Stacking classification

Input:$\mathrm{training}\mathrm{data} \mathrm{D}={\{x_i, y_i\}}_{i=1}^m$ Output: a stacking classifier H 1: Step 1: learn base-level classifiers 2: for t = 1 to T do 3: $\mathrm{learn} {\mathrm{h}}_{\mathrm{t}}$ based on D 4: end for 5: Step 2: construct new data set of predictions 6: for i =1 to m do 7: ${\mathrm{D}}_{\mathrm{h}}=\left\{{\mathrm{x}}_{\mathrm{i}}{}^{\prime },{ \mathrm{y}}_{\mathrm{i}}\right\}$$,\mathrm{where}{\mathrm{x}}_{\mathrm{i}}{}^{\prime }=\left\{{\mathrm{h}}_1\left({\mathrm{x}}_{\mathrm{i}}\right),\dots \mathrm{\dots },{\mathrm{h}}_{\mathrm{T}}\left({\mathrm{x}}_{\mathrm{i}}\right)\right\}$ 8: end for 9: Step 3: learn a meta-classifier 10: $\mathrm{learn}\mathrm{H} \mathrm{based} \mathrm{on} {\mathrm{D}}_{\mathrm{h}}$ 11: return H