Title: Optimal model selection in binary predictive data science models
Authors: Silvia Figini - University of Pavia (Italy)
Marta Galvani - University of Pavia (Italy) [presenting]
Abstract: A novel methodology is presented to assess predictive models for a binary target. One of the main weakness of the criteria proposed in the literature is not to take the financial costs of a wrong decision into account. The objective is to improve model assessment and selection. The methodological proposal can be of interest for a wide range of applications.We describe how our proposal performs in a real application in credit risk.