Survey on Lazy Ensemble Methods for Improving Accuracy of Lazy Learner?

In classification, to handle test instances, supervise learning is divided in to two parts Eagar and Lazy learning. Eagar learning build decision theory form training instance and that theory is applied on all test instances. Where Lazy learning is focus on each test instance and provides local optimal solution for each test instance. Diversity is the one issue of lazy learning, it suffers from reduce diversity. It is key issue in combination approach. It is possible to merge lazy and ensemble approach for build lazy learner. Lazy ensemble is build using ensemble approach that uses multiple classifiers and combines its predictions using combination methods and find optimal label. Many approaches for ensemble like lazy stacking, lazy bagging and other methods. Stacking with Lazy learning is used for building Lazy ensemble learner provides desire accuracy. Lazy Stacking Ensemble uses different individual “type” of classifiers as base classifiers for labelling new instance. Survey can be done only on numeric datasets and Lazy stacking is outperforming then other ensemble methods when it compared with other alternative methods in terms of classification accuracy that concern with diversity