Relevance Rank for Cluster Rules

A cluster rule has an antecedent (IF clause) and a consequent (THEN clause). The antecedent of a rule describes the clustering bounding box. The consequent encodes the cluster ID for the cluster described by the rule. Attributes in the rule antecedent are selected and ranked according to how well they distinguish the records in the cluster from the overall population. This ranking is the Relevance Rank for the attribute. A rank of 1 indicates the most important attribute. Attributes that distinguish records in the same way have equal Relevance Rank values. Only Show Attributes With Minimum Relevance Rank of R specifies that only those attributes with a relevance ranking less or equal to R are included in the rules antecedent. Specifying a minimum relevance rank controls the complexity of the rules by including only the most important attributes in the rule.

The default is to not check Only Show Attributes With Minimum Relevance Rank.

When a relevance rank is specified, the default Minimum Relevance Rank is 10; in this case, all rules with a relevance rank <= 10 are displayed.