A Simple and Effective Model-Based Variable Importance Measure
This addresses the need for interpretability in machine learning by providing a consistent method to understand feature influence, which is incremental as it builds on existing model-free approaches.
The paper tackles the challenge of quantifying predictor importance in supervised learning models, proposing a standardized model-based approach that works across various algorithms, including those like naive Bayes and support vector machines that lack built-in importance measures, and demonstrates its effectiveness through simulated and real data examples.
In the era of "big data", it is becoming more of a challenge to not only build state-of-the-art predictive models, but also gain an understanding of what's really going on in the data. For example, it is often of interest to know which, if any, of the predictors in a fitted model are relatively influential on the predicted outcome. Some modern algorithms---like random forests and gradient boosted decision trees---have a natural way of quantifying the importance or relative influence of each feature. Other algorithms---like naive Bayes classifiers and support vector machines---are not capable of doing so and model-free approaches are generally used to measure each predictor's importance. In this paper, we propose a standardized, model-based approach to measuring predictor importance across the growing spectrum of supervised learning algorithms. Our proposed method is illustrated through both simulated and real data examples. The R code to reproduce all of the figures in this paper is available in the supplementary materials.