The Kernel Density Integral Transformation
This work addresses feature preprocessing for machine learning and statistical methods on tabular data, offering a flexible alternative to existing methods, though it is incremental in nature.
The authors tackled the problem of feature preprocessing for tabular data by proposing the kernel density integral transformation, which subsumes linear min-max scaling and quantile transformation as limiting cases and can outperform them with tuning, achieving improved performance in tasks like correlation analysis and univariate clustering.
Feature preprocessing continues to play a critical role when applying machine learning and statistical methods to tabular data. In this paper, we propose the use of the kernel density integral transformation as a feature preprocessing step. Our approach subsumes the two leading feature preprocessing methods as limiting cases: linear min-max scaling and quantile transformation. We demonstrate that, without hyperparameter tuning, the kernel density integral transformation can be used as a simple drop-in replacement for either method, offering protection from the weaknesses of each. Alternatively, with tuning of a single continuous hyperparameter, we frequently outperform both of these methods. Finally, we show that the kernel density transformation can be profitably applied to statistical data analysis, particularly in correlation analysis and univariate clustering.