Visualizing Variable Importance and Variable Interaction Effects in Machine Learning Models

2021/11/27 by Alan Inglis, Alan N. Inglis, Andrew Parnell +2 · 3 citations
Computer Science · Psychology · #Data Analysis with R #Data Visualization and Analytics #Mental Health Research Topics

paper · pdf · doi:10.1080/10618600.2021.2007935

openalex publication_date 2021/11/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Variable importance, interaction measures, and partial dependence plots are important summaries in the interpretation of statistical and machine learning models. In this article, we describe new visualization techniques for exploring these model summaries. We construct heatmap and graph-based displays showing variable importance and interaction jointly, which are carefully designed to highlight important aspects of the fit. We describe a new matrix-type layout showing all single and bivariate partial dependence plots, and an alternative layout based on graph Eulerians focusing on key subsets. Our new visualizations are model-agnostic and are applicable to regression and classification supervised learning settings. They enhance interpretation even in situations where the number of variables is large. Our R package vivid (variable importance and variable interaction displays) provides an implementation. Supplementary files for this article are available online.

Citations

• Multivariate Adaptive Regression Splines
• Unbiased Recursive Partitioning: A Conditional Inference Framework
• Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation
• A Simple and Effective Model-Based Variable Importance Measure
• Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries

Cited by

• Adaptive Physics-Informed Neural Networks with Multi-Category Feature Engineering for Hydrogen Sorption Prediction in Clays, Shales, and Coals
• midr: Learning from Black-Box Models by Maximum Interpretation Decomposition
• Global Pattern and Disease Heterogeneity Drivers in Aging Populations