K-means clustering

[wby920920]

Hello, can you explain the meaning of the resting state cluster diagram, why the ratio is greater than 1? How are DMN and other networks determined? Actually, I just don’t understand Figure 3 and Figure 6

[egyptdj]

The k-means clustering model is trained with uniformly sampled FC matrices across time, and inferred with attended FC matrices within a subject.
There are three ratios provided in the paper, which are (i) female attended over uniform sampled, (ii) male attended over uniform sampled, and the (iii) (i) over (ii). The ratio (i) represents the clusters that the model attended more than by chance for female subjects, and the ratio (ii) represents same thing for male subjects. Accordingly, the ratio (iii) represents the clusters that the model attended more for female subjects than the male subjects.
If a certain pattern of FC connectivity was attended by the model more frequently for female subjects than for male subjects, then the cluster representing the pattern can have ratio (iii) over 1.0.

[wby920920]

Why are the ratios in (1) and (2) greater than 1?