A Novel Incremental Framework for Building Classifier Using Constraint Class Association Rules

Associative classification (AC) performs much better than other traditional classifiers. It generates a huge number of class association rules (CARs). Since users are interested in the subset rules, constraints introduced generation CARs. Real-world databases record-based which data is continuously added demands incremental mining. Hence, constraint (CCAR) mined from data. To limit and to remove duplicate redundant rule pruning techniques applied. improve accuracy classifier, selection using principality metric has been applied classifier constructed with possessing high principality. Then, evaluated single multiple prediction methods proposed measured. Experimental results show that relatively higher when compared algorithms.