Skip to main content

Table 2 The meaning of parameter setting

From: Effective hybrid feature selection using different bootstrap enhances cancers classification performance

Parameter

Value

Definition

NRuns

20

No of runs

Problem Dimensions

–

No of features in the dataset.

X*

2916

The number of data produced after the bootstrap resamples method.

M

100

The number of trees using in the Random Forest algorithm.

Criterion

–

The method which measures the quality of split, Entropy is applied.

min_samples_leaf

100

The minimum number of samples required to be at a leaf node.

RFE estimators

–

A supervised learning algorithm. LR is applied.

C

0.05

Regularization parameter.

Max-iteration

100

Max iteration in LR classifier.

Tol

0.0001

Tolerance to stop criteria in LR classification.

CV

10

No of folds in cross-validation.