A criterion for comparing two sets of steganalytical methods with a determined reliability

The paper is devoted to comparing two sets of steganalytical methods. It is assumed that one of the sets has been studied, i.e. it is known whether the results of the methods included in it are correlated. Checking the method's set for correlation is a complex computational task, so it makes sense to compare the sets as a whole. If it turns out that the sets are close, then the unknown set can be considered studied. The comparison is carried out with a determined reliability because it is important for using the results in the steganalytical forensics.