Simplifying Forest-based model into Decision Trees for Enhanced explainability

• Grant number: 24/00495-4
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2024
• End date: April 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: André Carlos Ponce de Leon Ferreira de Carvalho
• Grantee: Pedro Fernandez Tonso
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The main objective of this project is the development of a methodology capable of explaining how predictive models make their predictions. For such, we propose to simplify a complex ensemble predictor, such as the model induced by the Random Forest algorithm, into a single decision tree while preserving most of the relevant learned information and predictive power.