Random subspace and random projection nearest neighbor ensembles for high dimensional data
作者：
Highlights：
• A comparative study of nearest neighbor ensembles formed with random subsets and random projections.
• The best method depends on the type of high-dimensional data sets.
• Data complexity measures provide what relative performance could be expected for specific data sets.
摘要
•A comparative study of nearest neighbor ensembles formed with random subsets and random projections.•The best method depends on the type of high-dimensional data sets.•Data complexity measures provide what relative performance could be expected for specific data sets.
论文关键词：Nearest neighbor ensemble,High dimensional data,Random subspace method,Random projection method
论文评审过程：Received 11 March 2020, Revised 13 September 2021, Accepted 10 October 2021, Available online 14 November 2021, Version of Record 9 December 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.116078
