Attentive Hierarchical ANFIS with interpretability for cancer diagnostic

作者：

Highlights：

• The AH-ANFIS combines fuzzy inference in a hierarchical architecture.

• The hierarchical structure in fuzzy modeling overcome the rule explosion problem.

• Interpret the system’s conclusions based on fuzzy rules extraction.

• An EA is used to decompose the input space into the optimal input subsets.

• Improve the interpretability by pruning fuzzy if-then rules.

摘要

•The AH-ANFIS combines fuzzy inference in a hierarchical architecture.•The hierarchical structure in fuzzy modeling overcome the rule explosion problem.•Interpret the system’s conclusions based on fuzzy rules extraction.•An EA is used to decompose the input space into the optimal input subsets.•Improve the interpretability by pruning fuzzy if-then rules.

论文关键词：Neuro-fuzzy network,Attention,ANFIS,Interpretable AI,Cancer diagnostic,Hierarchical architecture

论文评审过程：Received 2 December 2020, Revised 27 December 2021, Accepted 28 March 2022, Available online 9 April 2022, Version of Record 20 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117099