Classification rule mining and association rule mining are
two important data mining techniques. Classification rule
mining aims to discover a small set of rules in the database
to form an accurate classifier (e.g., Quinlan 1992; Breiman
et al 1984). Association rule mining finds all rules in the
database that satisfy some minimum support and minimum
confidence constraints (e.g., Agrawal and Srikant 1994).
For association rule mining, the target of mining is not predetermined,
while for classification rule mining there is
one and only one pre-determined target, i.e., the class.
Both classification rule mining and association rule mining
are indispensable to practical applications. Thus, great
savings and conveniences to the user could result if the two
mining techniques can somehow be integrated. In this
paper, we propose such an integrated framework, called
associative classification. We show that the integration can
be done efficiently and without loss of performance, i.e.,
the accuracy of the resultant classifier.