Skip to main content

Table 1 The mean number of finally selected PCSNPs by each of the five feature selection methods and their resulting accuracy as well as the accuracy provided by using the entire swine SNPs in the data set

From: Discovery of significant porcine SNPs for swine breed identification by a hybrid of information gain, genetic algorithm, and frequency feature selection technique

Method

PCSNPs

Accuracy (%)

 

Linear

RBF

Linear

RBF

Whole SNPs

10,210

10,210

95.66±1.28

95.66±1.28

 

(100%)

(100%)

  

IG+GA

207.70 ±42.71

209.90 ±37.57

95.27 ±1.57

94.88 ±1.11

 

(2.03%)

(2.06%)

  

IG+modified GA

319.10 ±104.02

310.70 ±84.89

95.74 ±1.47

95.35 ±1.42

 

(3.13%)

(3.04%)

  

IG

413.30 ±88.22

410.30 ±88.22

95.43 ±1.53

95.58 ±1.51

 

(4.05%)

(4.05%)

  

IG+FFS

240.80 ±15.33

240.80 ±15.33

95.58 ±1.27

95.66 ±1.22

 

(2.36%)

(2.36%)

  

IG+modified GA+FFS

164.90±36.11

164.90±36.11

94.81 ±1.46

95.12 ±1.55

 

(1.62%)

(1.62%)