Table 3 Clustering performance comparison.

From: Enhancing education quality with hybrid clustering and evolutionary neural networks in a multi phase framework

 

K-means

Fuzzy clustering

HPEFCM-FSP

Correctly clustered instance

40

47

52

Incorrectly clustered instance

18

11

6

Silhouette score

0.65

0.72

0.80

Dunn index

0.25

0.30

0.35

Clustering accuracy

68.97%

81.03%

89.66%

Area under the ROC curve (AUC)

0.70

0.75

0.85

Back to article page