Multivariate multi-layer classifier

作者：

Highlights：

• Multi-layer classifiers (MLC) are simpler straight-trunk decision trees.

• Theoretical foundation is provided for building MLC with binary and ternary splits.

• MLC structure is further condensed by integrating oblique discriminant hyperplanes.

• Proposed decision trees are proven simpler with better classification performance.

摘要

•Multi-layer classifiers (MLC) are simpler straight-trunk decision trees.•Theoretical foundation is provided for building MLC with binary and ternary splits.•MLC structure is further condensed by integrating oblique discriminant hyperplanes.•Proposed decision trees are proven simpler with better classification performance.

论文关键词：Classification,Classifiers,Multivariate decision tree,Machine learning,Tree construction

论文评审过程：Received 28 July 2021, Revised 15 May 2022, Accepted 9 July 2022, Available online 11 July 2022, Version of Record 17 July 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108896