This table provides the analysis of dependencies between items stats. If there is a dependency, for each item, the efficiency of one characteristic regarding the other is estimated as a difference between real and expected values. If there is no dependency, items are grouped into clusters based on similar values of characteristics