Skip to main content

Table 3 Classification performance on test data set for dictionary size = 60

From: Iris classification based on sparse representations using on-line dictionary learning for large-scale de-duplication applications

Class

Residual parameter

 

0.5

0.05

0.005

Class-1 (Stream)

90.5

97

93.83

Class-2 (Flower)

91.18

94.12

88.2

Class-3 (Jewel-Shaker)

100

100

100

  1. Boldface data represents the best performance.