Skip to main content

Table 3 Accuracy of stress according to different machine learning algorithms

From: Machine learning to reveal an astute risk predictive framework for Gynecologic Cancer and its impact on women psychology: Bangladeshi perspective

Model

Tool

AUC

CA

F1

Precision

Recall

Specificity

Support

SVM

Orange

0.746

0.744

0.698

0.736

0.744

0.462

Sklearn

0.615

 

0.6

0.59

0.62

39

Random

Orange

0.746

0.725

0.707

0.705

0.725

0.525

Forest

Sklearn

0.692

 

0.7

0.73

0.69

39

Logistic

Orange

0.796

0.769

0.752

0.757

0.769

0.579

Regression

Sklearn

0.769

 

0.77

0.76

0.77

39

AdaBoost

Orange

0.744

0.788

0.785

0.783

0.788

0.683

Sklearn

0.79

 

0.79

0.79

0.79

39

Naïve

Orange

0.753

0.763

0.763

0.763

0.763

0.672

Bayes

Sklearn

0.785

 

0.78

0.79

0.78

39

Neural

Orange

0.767

0.75

0.747

0.745

0.75

0.631

Network

Sklearn

0.692

 

0.57

0.48

0.69

39

kNN

Orange

0.646

0.669

0.627

0.618

0.669

0.394

Sklearn

0.615

 

0.6

0.59

0.62

39

CN2 rule

Orange

0.721

0.694

0.696

0.699

0.694

0.595

Inducer

Sklearn

 

 

 

Decision

Orange

0.7

0.713

0.717

0.723

0.713

0.639

Tree

Sklearn

0.692

 

0.7

0.71

0.69

39

Quadratic

Orange

 

 

 

Classifier

Sklearn

0.744

 

0.7

0.75

0.74

39