Adjusting and generalizing CBA algorithm to handling class imbalance

Associative classification has attracted substantial interest in recent years and been shown to yield good results. However, research in this field tends to focus on the development of class classifiers, but the required probability classifier of imbalance data has not been addressed comprehensively. This investigation presents a new associative classification method called Probabilistic Classification based on Association Rules (PCAR). PCAR is based on modifying the rule sorting index, the pruning method, and the scoring procedure in the CBA algorithm. CBA can be generalized to construct a probability classifier. Additionally, it can improve the efficiency of associative classification for predicting imbalance data. Experiments that use both benchmarking datasets and real-life application datasets reveal that the new method outperforms the previous associative classification algorithm and C5.0 for all datasets. Also, in some datasets, the predictive performance exceeds that achieved by logistic regression and the use of a neural network.