Skip to main content

Table 1 Percent true classification of chromatographic data by discriminatory models

From: Data mining of plasma peptide chromatograms for biomarkers of air contaminant exposures

Treatment Groups

Expected Classification

Dataset

Percent True Classification by Models

   

GA

SVM

Clean air, 0 h and 24 h recovery post-exposure

Clean air

dewarped

100.00

100.00

  

non-dewarped

91.67

83.33

EHC-93, 0 h and 24 h recovery post-exposure

EHC-93

dewarped

100.00

83.33

  

non-dewarped

66.67

83.33

0 h recovery post- exposure (Clean air and EHC-93)

0 h

dewarped

91.67

100.00

  

non-dewarped

75.00

91.67

24 h recovery post- exposure (Clean air and EHC-93)

24 h

dewarped

91.67

91.67

  

non-dewarped

75.00

91.67

  1. Percent true classification of all chromatographic data (dewarped or non-dewarped) into exposure (clean air and EHC-93) and post-exposure recovery time (0 h, 24 h) groups by discriminatory models based on the genetic algorithm (GA) and the support vector machine algorithm (SVM).