Ensemble learning is an important element in machine learning. However, two essential tasks, including training base classifiers and finding a suitable ensemble balance for the diversity and accuracy of these base classifiers, are need to be achieved. In this paper, a novel ensemble method, which utilizes a multimodal multiobjective differential evolution (MMODE) algorithm to select feature subsets and optimize base classifiers parameters, is proposed. Moreover, three methods including minimum error ensemble, all Pareto sets ensemble, and error reduction ensemble are employed to construct ensemble classifiers for executing classification tasks. Experimental results on several benchmark classification databases evidence that the proposed algorithm is valid.
CITATION STYLE
Wang, J., Wang, B., Liang, J., Yu, K., Yue, C., & Ren, X. (2020). Ensemble Learning via Multimodal Multiobjective Differential Evolution and Feature Selection. In Communications in Computer and Information Science (Vol. 1159 CCIS, pp. 439–453). Springer. https://doi.org/10.1007/978-981-15-3425-6_34
Mendeley helps you to discover research relevant for your work.