Skip to main content

Table 4 Comparing clustering algorithms (Left to Right) Recognition rates for: independently applied k-means, independent Gaussian mixture models (i-mix), alternating row and column hard assignment row-column mixtures (rc-mix), sequentially optimized row-column mixtures (s-rc-mix) and row-column mixtures optimized using our variational method (v-rc-mix). We recover the block constant structure of synthetic data consisting of 50 × 50 element matrices with 5 row and 5 column classes.

From: Analyzing in situ gene expression in the mouse brain with image registration, feature extraction and block clustering

  k-means i-mix rc-mix s-rc-mix v-rc-mix
recognition rate .71 ± .04 .72 ± .04 .77 ± .04 .83 ± .04 .84 ± .05
time (sec.) .15 ± .02 .22 ± .01 1.0 ± .06 2.3 ± .2 110 ± 10