To enhance the reliability of the LS-SVM model, it
is necessary to preprocess the modeling data. Due to the
large difference in the order of magnitude of the value,
the available samples are scaled into 0-1 using a
normalization preprocessing method. Different from the
general normalization method, formulated as in (12), a
logarithmic normalization method is utilized. The
logarithmic method can deal with a wider range of data
for the unbalanced data with a better performance than
the general normalization method. The logarithmic
normalization function can be formulated in (13).