Optimization of stacking ensemble Configuration based on various metahueristic algorithms

Stacking Ensemble is a collective frame work having strategies to combine the predictions of learned classifiers to generate predictions as new instances occur. In early research it has been proved that a stacking ensemble is usually more accurate than any other single-component classifier. Many ensemble methods are proposed, but still it is a difficult task to find the suitable ensemble configuration. Meta-heuristic methods can be used as a solution to find optimized configurations. Genetic algorithms, Ant Colony algorithms are some popular approaches on which current researches are going on. This paper is about meta-heuristic approaches used so far for the optimization of stacking configuration and what work can be done in the future to overcome the shortcomings of existing techniques. Particle swarm optimization based stacking ensemble framework can be applied to get better results. A number of studies, comparison and experiments are presented by extracting from a large no of references.