2.3.3. QUEST Algorithm
QUEST stands for a quick, unbiased, efficient statistical tree and is a
relatively new algorithm for binary tree increasing. This deals with
split field selection and different split-point selection. In Search,
the univariate split conducts roughly unbiased field selection. If all
the predictor fields are equally informative concerning the target
field, QUEST selects any of the predictor fields with equal likelihood.
QUEST offers many of Classification and Regression Trees’ (C&RT)
benefits. Still, the trees can become unusable, like C&RT. Automatic
cost-complexity pruning can be applied to the QUEST tree to reduce its
scale. Concerning this study, the boosting method was used to build the
QUEST model to enhance accuracy (11). The hyperparameters of the SVM
algorithm were maximum tree depth of 5, maximum surrogates of 5, minimum
records in parent branch of 2%, minimum records in child branch of 1%,
and the number of component models for boosting was ten. The model
hyperparameters were optimized by SMO approximation (13).