Skip to main content

Table 3 The performance evaluation of the wrapper feature selection methods

From: A comparison study on feature selection of DNA structural properties for promoter prediction

Classifier

Wrapper

Specificity

Sensitivity

F-measure

ROC score

Dimensionality

SVM

SVM

0.761

0.723

0.742

0.786

320

KNN

KNN

0.705

0.731

0.718

0.745

300

SVM

NULL

0.733

0.72

0.731

0.732

All

KNN

NULL

0.637

0.753

0.69

0.723

All

  1. The performance comparison of two classifiers (SVM and kNN) for promoter classification: with the wrapper feature selection (the first two rows) vs. without wrapper feature selection (the last two rows).
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us