202502012204
tags: #machine-learning #preprocessing #features

Feature scaling normalizes input features to similar ranges, preventing features with larger scales from dominating the learning process.

Why it matters:

• Features like house size (1000-5000 sq ft) vs. bedrooms (1-5) have vastly different scales
• Gradient Descent converges faster with scaled features
• Distance-based algorithms (k-NN, clustering) work better with normalized data

Common scaling methods:

• Normalization (Min-Max): Scales to [0,1] range
• Standardization (Z-score): Mean=0, standard deviation=1
• Robust scaling: Uses median and interquartile range

Apply the same scaling parameters from training data to test data to avoid Data Distribution Mismatch.

Feature scaling is particularly important for algorithms using Regularization since it ensures penalties are applied fairly across all features.

Reference

Machine Learning Yearning by Andrew Ng