Table 4 Results for second stage clustering. Dataset \(D_1\) (N = 2,768). SIL silhouette, DB Davies-Bouldin, CH Calinski-Harabasz. Best metric value achieved appears in bold
From: Machine learning based study for the classification of Type 2 diabetes mellitus subtypes
Obs. per group | Intrinsic index | ||||||
|---|---|---|---|---|---|---|---|
Strat. | 0 | 1 | 2 | 3 | SIL | DB | CH |
2.1 | 1160 | 523 | 511 | 574 | 0.1976 | 1.5518 | 629.61 |
2.2 | 1140 | 555 | 542 | 531 | 0.1883 | 1.6004 | 607.14 |
2.3 | 1138 | 546 | 511 | 573 | 0.1959 | 1.5530 | 631.39 |
2.4 | 1092 | 584 | 552 | 540 | 0.1798 | 1.6188 | 601.69 |
2.5 | 1155 | 530 | 549 | 534 | 0.2118 | 1.4701 | 671.79 |
2.6 | 1196 | 488 | 452 | 632 | 0.2580 | 1.2120 | 934.23 |