Skip to main content
Figure 10 | BMC Bioinformatics

Figure 10

From: Finding reproducible cluster partitions for the k-means algorithm

Figure 10

Performance chart for 1,000 dataset. Ordered Cramérs' V for 100 runs of the SeCo framework with selection of the best solution being (a) SSQ only and (b) Median Internal Cramérs' V for the top 10% of solutions. Results for the dataset with 1,000 data points. Red lines highlights k = 10, blue lines k = 9 and the black lines k = 8. Confidence intervals for k = 8 (Upper: +0.018, Lower: -0.029) k = 9 (Upper: +0.017, Lower: -0.027) k = 10 (Upper: +0.016, Lower: = -0.026).

Back to article page