Ordinal classification with a spectrum of information sources

作者：

Highlights：

• Different types of information are considered for ordinal classification.

• Labeling rules and learned distance metrics are combined into different methods.

• Using frequentist relative information outperforms using relative information.

摘要

•Different types of information are considered for ordinal classification.•Labeling rules and learned distance metrics are combined into different methods.•Using frequentist relative information outperforms using relative information.

论文关键词：Ordinal classification,Absolute information,Relative information,Frequentist relative information,Distance metric learning

论文评审过程：Received 26 May 2021, Revised 16 June 2022, Accepted 10 July 2022, Available online 18 July 2022, Version of Record 26 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118163