The paper proposes a new GEP-based batch ensemble classifier constructed using the stacked generalization concept. In our approach combination of base classifiers involves evolving the meta-gene using genes induced by GEP from randomly generated combinations of instances with randomly selected subsets of attributes. The main property of the discussed classifier is its scalability allowing adaptation to the size of the dataset under consideration. To validate the proposed classifier, we have carried-out computational experiment involving a number of publicly available benchmark datasets. Experiment results show that the approach assures good performance, scalability and robustness.
CITATION STYLE
Jȩdrzejowicz, J., & Jȩdrzejowicz, P. (2017). Gene Expression Programming Ensemble for Classifying Big Datasets. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10449 LNAI, pp. 3–12). Springer Verlag. https://doi.org/10.1007/978-3-319-67077-5_1
Mendeley helps you to discover research relevant for your work.