Evolutionary bagging for ensemble learning

Ensemble learning has gained success in machine with major advantages over other methods. Bagging is a prominent ensemble method that creates subgroups of data, known as bags, are trained by individual methods such decision trees. Random forest example bagging additional features the process. Evolutionary algorithms have been for optimisation problems and also used learning. gradient-free work population candidate solutions maintain diversity creating new solutions. In conventional bagged learning, bags created once content, terms training examples, fixed our paper, we propose evolutionary where utilise to evolve content order iteratively enhance providing bags. The results show outperforms (bagging random forests) several benchmark datasets under certain constraints. We find can inherently sustain diverse set without reduction performance accuracy.