To do or not to do? Cost-sensitive causal classification with individual treatment effect estimates

Wouter Verbeke, Diego Olaya, Marie-Anne Guerry, Jente Van Belle

Abstract: Highlights•We introduce a cost-sensitive decision-making framework for causal classification.•We derive the classification boundary to maximize the expected causal profit.•We establish a new cost-sensitive ranking approach with individual treatment effects.•Experiments demonstrate an increase in profit compared to cost-insensitive ranking.