.png){kind=icon}
.png){kind=icon}
Return
A visualization-driven decision support system for selecting feature attribution methods
.png){kind=icon}
PRE
AI
DOI：10.1016/j.is.2025.102661.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Introduce Explainalytics, an interactive visual analytics tool for comparing and selecting feature attribution-based machine learning model explanation methods. • Integrate quantitative evaluation metrics with linked visualizations to support human-centered interpretation. • Demonstrate a within-subject user study (n=10) showing Explainalytics significantly reduces cognitive workload and increases usability versus the baseline, and two case studies illustrating how practitioners use Explainalytics to compare attribution methods and explore fairness-related aspects of explanations.
Journal
I
IF:
3.4
Papers:
109
Citations:
0