Interpretable kNN (ikNN) is an interpretable classifier that provides explanations for its predictions. It uses an ensemble of 2d kNN models to increase interpretability while maintaining accuracy. The ikNN model weighs the influence of input features by considering their predictive power and consistency.

10m read timeFrom towardsdatascience.com
Post cover image
Table of contents
Interpretable kNN (ikNN)Standard kNN ClassifiersThe ikNN AlgorithmExampleVisualizationsAccuracy TestsConclusions

Sort: