A parameter randomization approach for constructing classifier ensembles

作者：

Highlights：

• We propose a novel randomization-based approach for classifier ensemble construction.

• It samples the parameters of the base classifiers from a pre-defined distribution.

• As an example we derive the parameter distribution of some linear bagged classifiers.

• We then simulate bagging by using the derived distribution.

摘要

•We propose a novel randomization-based approach for classifier ensemble construction.•It samples the parameters of the base classifiers from a pre-defined distribution.•As an example we derive the parameter distribution of some linear bagged classifiers.•We then simulate bagging by using the derived distribution.

论文关键词：Multiple classifier systems,Ensemble construction techniques,Randomization,Bagging

论文评审过程：Received 23 June 2016, Revised 3 February 2017, Accepted 26 March 2017, Available online 29 March 2017, Version of Record 6 April 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.03.031