Feature Importance Measurement based on Decision Tree Sampling

CHAO HUANG, Diptesh Das, Koji Tsuda

Keywords: Decision tree, Interpretability, Random Forest, SAT, Reliability, Trustworthy AI

Abstract: Random forest is effective for prediction tasks but the randomness of tree generation hinders interpretability in feature importance analysis. To address this, we proposed a SAT-based method for measuring feature importance in tree-based model. Our method has fewer parameters than random forest and provides higher interpretability and stability for the analysis in real-world problems.

Submission Number: 106