Abstract
SOME MEASURES RELATING PARTITIONS USEFUL FOR COMPUTATIONAL INTELLIGENCE We investigate a number of measures relating partitions. One class of measures we consider are congru- ence measures. These measures are used to calculate the similarity between two partitionings. We provide a number of examples of this type of measure. Another class of measures we investigate are prognosti- cation measures. This type of measure, closely related to a concept of containment between partitions, is useful in indicating how well knowledge of an objects class in one partitioning indicates its class in a second partitioning. We apply our measures to some data mining applications. One example is in choos- ing the appropriate level of a concept hierarchy. We also introduce a measure of the non-specificity of a partition. This measures a feature of a partition related to the granularity of the constituent classes of the partition.