Algorithm 2 Pseudocode of One-hot encoding, Min–Max Normalization and Z-score Standardization

Input: dataset features $\mathrm{F}$, class label $C$

Output: Pre-processed dataset

MinMaxScaler$(D, \mathrm{F}, i)$:

$X_{normalized}$ = 0

max $\leftarrow$ maximum value among all values of column $i\in \mathrm{F}$ in $D$

min $\leftarrow$ minimum value among all values of column $i\in \mathrm{F}$ in $D$

$X_{normalized}\leftarrow \frac{X-X_{min\_value}}{X_{max\_value}-X_{min\_value}}$ // Equation (1)

Return $X_{normalized}$

Standardize$(D,\mathrm{F}, i)$:

$X_{normalized}$ = 0

$\mu \leftarrow \mathrm{mean}\mathrm{value}\mathrm{of}\mathrm{column}i\in \mathrm{F}\mathrm{in}D$

$\sigma \leftarrow$ $\mathrm{standard}\mathrm{deviation}\mathrm{value}\mathrm{of}\mathrm{column}i\in \mathrm{F}\mathrm{in}D$

$X_{normalized} \frac{X-\mu }{\sigma }$// Equation (2)

Return $X_{normalized}$

Begin:

$D’$$\leftarrow$ [ ] // Normalized/ Standardized dataset

$F$$\leftarrow$ Hot-encoding dataset $D$

For each item $i\in \mathrm{F}$ in $\left(D\right)$ do:

$D’$$\leftarrow$$\mathrm{MinMaxScaler}(D, \mathrm{F}, i)$ // both min-max and z-score method is

$D’$$\leftarrow$$\mathrm{Standardize}(D, \mathrm{F}, i)$ // executed separately

End For

End