.png){kind=icon}
.png){kind=icon}
Return
Knowledge-guided counterfactual explanations for diabetes risk decision support: A directional intervention taxonomy
.png){kind=icon}
PRE
AI
V
DOI：10.1016/j.ijmedinf.2026.106555.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Counterfactuals made actionable via a 5-class feature intervention taxonomy. • Taxonomy-consistent actionability reaches 0.9880 with eliminated wrong-direction violations. • Rule-based translation improves actionability from 0.6655 to 0.9880. • Per-query monotonicity enforces taxonomy-defined intervention directionality. • A lightweight clinical knowledge layer and rule-based translator automate per-query constraints.
Journal
IF:
4.1
Papers:
4.5K
Citations:
1.1W
Organization
Cited Papers