A novel ensemble method for k-nearest neighbor

作者：

Highlights：

• We proposed a weighted heterogeneous distance metric (WHDM).

• We presented WHDM and Dempster–Shafer theory based kNN algorithm.

• We proposed a multimodal perturbation method (RRSB) for kNN ensemble.

• The effectiveness of our algorithms was shown on multiple UCI data sets and a KDD data set.

摘要

•We proposed a weighted heterogeneous distance metric (WHDM).•We presented WHDM and Dempster–Shafer theory based kNN algorithm.•We proposed a multimodal perturbation method (RRSB) for kNN ensemble.•The effectiveness of our algorithms was shown on multiple UCI data sets and a KDD data set.

论文关键词：Distance metric,k-nearest neighbor,Ensemble learning,Random subspace,Evidence theory

论文评审过程：Received 10 April 2018, Revised 19 June 2018, Accepted 1 August 2018, Available online 2 August 2018, Version of Record 8 August 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.08.003