Data mining method and apparatus using rate of common records as a measure of similarity

摘要

The present invention generates a plurality of rules comprising a consequence including records having the characteristics of the target attribute and record items and a precedence including at least one record item in an attribute appearance area decided by at least one record item in a database comprising a plurality of records having at least one record item for the database, calculates the similarity between the rules decided by the rate of records common to two rules among the plurality of rules among respective records satisfying the consequences, and obtains inclusive relations between the rules from the rate of matched record items to mismatched record items in the respective attribute appearance areas of the two rules.