A Dynamic Weighted Ensemble to Cope with Concept Drifting Classification

In the real world concepts are not stable and change with time and a lot of other hidden factors. Stream classifiers should be sensitive to the drifting of concept in an automatic way. In this paper, we proposed a new weighted majority strategy for the ensemble classifier. We periodically created and evaluated component classifiers that constitute the ensemble then we used the weighted ensemble to make global prediction. We empirically evaluated two kinds of concept drifting: the SEA concept drifting and the moving hyper-plane problem. Experiment results showed that our proposed method was very effective to deal with concept drifting.