Associative classification method applies association rule mining technique in classification and achieves higher classification accuracy. However, it is a known fact that associative classification typically yields a large number of rules, from which a set of high quality rules are chosen to construct an efficient classifier. Hence, generating, ranking and selecting a small subset of high-quality rules without jeopardizing the classification accuracy is of prime importance but a challenging task indeed. This paper proposes lazy learning associative classification method, which delays processing of the data until a new sample needs to be classified. This proposed method is useful for applications where the training dataset needs to be frequently updated. Experimental results show that the proposed method outperforms the CBA method. © 2011 Springer-Verlag.