# Occam's razor is a principle that suggests selecting the simplest solution that sufficiently explains the data.
In data science, this translates to favoring simpler models (fewer features) when performance is similar.
Therefore, the model with the fewest features and the highest performance is preferred - balancing simplicity and effectiveness.
Why the other options are incorrect:
* B: Poor performance undermines utility.
* C & D: More features add complexity and risk overfitting, making them less desirable when simpler models suffice.
Official References:
* CompTIA DataX (DY0-001) Official Study Guide - Section 3.2:"Simplicity in models improves interpretability and robustness. When models perform similarly, the simpler model should be preferred."
* Data Science Principles, Chapter 5:"Occam's razor encourages the use of fewer features to minimize complexity while preserving accuracy."
-