IEEE VIS 2025 Content: XplainAct: Visualization for Personalized Intervention Insights

XplainAct: Visualization for Personalized Intervention Insights

Yanming Zhang -

Krishnakumar Hegde -

Klaus Mueller -

Screen-reader Accessible PDF
Download camera-ready PDF
Image not found
1.Data Scientists and Machine Learning Engineers Application: They can integrate XplainAct’s ideas into their own pipelines to interpret and validate models more effectively, especially in applications where local explanations are as important as global patterns. 2.Public Health Researchers and Epidemiologists Application: Use XplainAct to identify region-specific or subgroup-specific causal drivers and simulate policy interventions (e.g., improving mental health resources) in a way that accounts for local heterogeneity. 3.Political Analysts and Social Scientists Application: Campaign strategists or sociopolitical researchers can simulate policy changes or demographic shifts and assess how they might influence election outcomes in targeted regions.
Keywords

Explainable AI, Causality, Visual Analytics

Abstract

Causality helps people reason about and understand complex systems, particularly through what-if analyses that explore how interventions might alter outcomes. Although existing methods embrace causal reasoning using interventions and counterfactual analysis, they primarily focus on effects at the population level. These approaches often fall short in systems characterized by significant heterogeneity, where the impact of an intervention can vary widely across subgroups. To address this challenge, we present XplainAct, a visual analytics framework that supports simulating, explaining, and reasoning interventions at the individual level within subpopulations. We demonstrate the effectiveness of XplainAct through two case studies: investigating opioid-related deaths in epidemiology and analyzing voting inclinations in the presidential election.