➤ Why is feature scaling required?

Feature scaling is important in machine learning when models rely on the magnitude of features. Without scaling, features with larger ranges can dominate the learning process.

For example:

• In a dataset with age (0-100) and income (0–100,000), the model may give too much weight to income just because of its scale.
• Algorithms like KNN, SVM, Gradient Descent-based models, and PCA are sensitive to feature scales.

📌 “One feature overshadows other”

Exactly. This is what happens when you don’t scale: features with larger values overshadow smaller-range features, even if they’re equally important.

🧮 Types of Feature Scaling

There are several methods. You mentioned some:

1. Min-Max Scaling (0–1 Scaling)

x′=x−min⁡(x)max⁡(x)−min⁡(x)x' = \frac{x - \min(x)}{\max(x) - \min(x)}

• Transforms values to range [0, 1].
• Used when you want to preserve the shape of the original distribution.

✅ Pros: Preserves relationships

❌ Cons: Sensitive to outliers

2. Standardization (Z-score Normalization)