xxx List of Figures
18.8 Misleading clusters. . . . . . . . . . . . . . . . . . . . . . . 583
18.9 Patterns in different clusters can be similar. . . . . . . . . . 584
18.10 The k-means algorithm with k = 2. . . . . . . . . . . . . . 585
18.11 Different k-means results on the same data. . . . . . . . . . 586
18.12 Cluster quality asse ssment: diameter versus distance to other
clusters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 587
18.13 Cluster quality assessment: cluster radius. . . . . . . . . . . 588
18.14 A clustering of the top 3 5 genes from the leukemia data. . . 589
18.15 The residual-based cluster confidence approach. . . . . . . . 590
18.16 A hierarchical clustering of the yeast sporulation data. . . . 593
18.17 Linkage types