Back to results for “Fundamentals”
Machine Learning Fundamentals › Feature Selection Basics

What are common mistakes to avoid with Feature Selection Basics

Intermediate

Answer

A common mistake is selecting features using the entire dataset including the test set, which leaks information and gives misleading results.
Real-world example A hospital predicting patient risk may start with fifty medical measurements but use feature selection to narrow it down to the ten most useful ones, making the model faster and easier to explain to doctors.

Common follow-ups: What is feature engineering, What is dimensionality reduction, What is correlation