Skip to main content

Table 3 Classification Accuracy of different classifiers with selected features

From: Using multitask classification methods to investigate the kinase-specific phosphorylation sites

Methods

Window size

Feature number

aveAc

MetaPred

NA

NA

0.7997

LS-SVMs

7

560

0.7611

 

opt

560

0.7939

*MT-Feat3

7

25

0.7605

 

opt

23

0.7972

MTLS-SVMs

7

560

0.7595

 

opt

560

0.7936

*MTLS-SVMs

7

20

0.7621

 

opt

26

0.7962

#MTLS-SVMs

7

12

0.7455

 

opt

18

0.792

  1. Five fold cross validation and grid fitting of parameters are used to estimate the performance of all the classifiers.
  2. *Using features selected by the MT-Feat3 method.
  3. #Using features filtered by the Metric multi-dimensional scaling method.