Skip to main content

Table 7 Classification accuracy of the proposed method with and without LSM

From: Dissimilarity space reinforced with manifold learning and latent space modeling for improved pattern classification

Dataset

Proposed Algorithm

Proposed Algorithm without LSM

Accuracy(± std)

\({f}_{1}\)

Accuracy(± std)

\({f}_{1}\)

Iris

    

 Best (20,linSVM)

98.00(± 0.2)

0.98(± 0.03)

98.00(± 0.0)

0.96(± 0.03)

 Average

96.33(± 1.7)

0.95(± 0.03)

96.67(± 0.1)

0.96(± 0.03)

Glass

    

 Best(60,1-NN)

77.57(± 2.1)

0.56(± 0.32)

76.64(± 1.5)

0.50(± 0.30)

 Average

71.61(± 2.4)

0.50(± 0.3)

71.85(± 2.5)

0.42(± 0.60)

Ionosphere

    

 Best(50,5-NN)

95.73(± 1.1)

0.94(± 0.03)

91.45(± 1.3)

0.94(± 0.07)

 Average

93.38(± 1.5)

0.92(± 0.03)

93.16(± 0.2)

0.83(± 0.07)

Monk

    

 Best(60,polySVM2)

92.62(± 0.4)

0.89(± 0.00)

90.98(± 1.1)

0.87(± 0.00)

 Average

89.55(± 0.9)

0.87(± 0.01)

88.52(± 0.7)

0.86(± 0.01)

BreastCancer

    

 Best(50,11-NN)

94.20(± 0.8)

0.91(± 0.03)

94.20(± 0.1)

0.91(± 0.03)

 Average

91.83(± 0.7)

0.8(± 0.08)

91.40(± 0.6)

0.73(± 0.09)

Wine

    

 Best(10,linSVM)

74.72(± 1.2)

0.69(± 0.17)

75.28(± 0.5)

0.68(± 0.18)

 Average

74.37(± 1.8)

0.6(± 0.2)

73.53(± 1.1)

0.5(± 0.2)

USPS_nE

    

 Best(60,linSVM)

96.80(± 0.2)

0.89(± 0.17)

94.40(± 0.9)

0.90(± 0.11)

 Average

93.55(± 1.5)

0.83(± 0.2)

74.10(± 2.3)

0.85(± 0.17)

USPS_E

    

 Best(120,linSVM)

91.40(± 1.0)

0.86(± 0.09)

90.60(± 0.3)

0.80(± 0.12)

 Average

88.03(± 0.1)

0.78(± 0.11)

87.25(± 0.2)

0.79(± 0.14)

Digit5

    

 Best(10,linSVM)

99.50(± 0.1)

0.98(± 0.01)

99.50(± 0.0)

0.97(± 0.01)

 Average

99.19(± 0.1)

0.98(± 0.01)

99.06(± 0.1)

0.98(± 0.01)

MNIST

    

 Best(50,linSVM)

94.05(± 3.5)

0.85(± 0.02)

88.88(± 5.03)

0.75(± 0.03)

 Average

90.46(± 4.6)

0.83(± 0.02)

85.16(± 4.7)

0.73(± 0.02)