Genetic_Programming_Theory_and_Practice_XIII

Multiclass Classification Through Multidimensional Clustering 233

Ta b l e 3

Comparison between M2GP, M3GP and eM3GPHRT

IM-3

WAV

SEG

IM-10

YST

VOW

M-L

Training fitnessM2GP

89.4

98.2

87.4

96.8

91.4

62.6

95.9

100

M3GP

94.7

99.6

90.7

98.1

93.0

68.5

100

100

eM3GP

86.7

98.2

81.8

96.1

92.0

61.0

87.8

100

Test fitnessM2GP

80.2

93.8

84.9

95.6

90.2

53.8

85.9

63.0

M3GP

79.0

95.4

84.3

95.6

91.0

56.2

93.8

57.1

eM3GP

80.8

93.2

81.2

94.7

90.3

56.1

78.6

65.1

# NodesM2GP

37

24

126

43

117

146

49

33

M3GP

110

66

71

111

239

274

53

13

eM3GP

4

8

3

8

58

14

10

4

# DimensionsM2GP

3

(1–8)

2

(1–4)

5

(2–10)

4

(3–8)

7

(4–10)

6

(1–13)

9

(4–18)

10

(7–12)

M3GP

12

(1–17)

5

(2–8)

31

(29–37)

11

(5–21)

12

(11–16)

13

(11–18)

20

(16–20)

12

(10-13)

eM3GP

1

(1–4)

1

(1–5)

1

(1–10)

6

(2–10)

7

(3–12)

10

(1–16)

4

(1–14)

2

(1–11)

The values refer the medians of 30 runs. The best values are in bold (if more than one, it means there is nostatistically significant difference between the medians)Whenever a result is said to be significantly different (better or worse) from another, it means the difference isstatistically significant according to the Friedman test with Bonferroni-Holm correction using the 0.05 significancelevel