Semi-supervised extensions of multi-task tree ensembles

作者：

Highlights：

• Multi-criteria approaches improve the performance of semi-supervised decision trees.

• Scale inconsistencies can be effectively handled via proximity measures.

• Using target relations enhances predictive performance of semi-supervised decision trees.

摘要

•Multi-criteria approaches improve the performance of semi-supervised decision trees.•Scale inconsistencies can be effectively handled via proximity measures.•Using target relations enhances predictive performance of semi-supervised decision trees.

论文关键词：Semi-supervised learning,Multi-task learning,Multi-objective trees,Ensemble learning,Totally randomized trees

论文评审过程：Received 12 February 2020, Revised 11 October 2021, Accepted 15 October 2021, Available online 23 October 2021, Version of Record 14 November 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108393