A New Data Mining Method Using Organizational Coevolutionary Mechanism

Jing Liu, Weicai Zhong, Fang Liu, Licheng Jiao

Abstract: Organizational coevolutionary algorithm for classification (OCEC), is designed with the intrinsic properties of data mining in mind. OCEC makes groups of examples evolved, and then rules are extracted from these groups of examples at the end of evolution. OCEC is first compared with G-NET and JoinGA. All results show that OCEC achieves a higher predictive accuracy. Then, the scalability of OCEC is studied. The results show that the classification time of OCEC increases linearly.