Table 3 Utilizing p-values for feature selection in diabetes detection: a comparison of various FE techniques.

From: Generalizability of machine learning models for diabetes detection a study with nordic islet transplant and PIMA datasets

Feature selection

Data set

Microarray gene data set

PIMA dataset

FE techniques

ABC

ABC-PSO

ABC

ABC-PSO

Class

Diab

Non-Diab

Diab

Non-Diab

Diab

Non-Diab

Diab

Non-Diab

HS

P value

< 0.05

0.4086

0.2913

0.4187

0.3405

0.2929

0.2903

0.2715

0.3166

DFA

P value

< 0.05

0.3135

0.4998

0.4056

0.4798

0.4989

0.4997

0.4848

0.4892

EHA

P value

< 0.05

0.2364

0.4087

0.4112

0.4105

0.3908

0.1622

0.2075

0.2647

Back to article page