Supervised Learning

Supervised learning is like learning with a teacher. You're given examples with the correct answers, and you learn the pattern that connects inputs to outputs.

What Makes It "Supervised"?

The key characteristic: labeled data. Every training example includes both:

• Features (X): The input data
• Label (y): The correct answer

The model learns to map features to labels, then predicts labels for new data.

Two Types of Supervised Learning

1. Regression: Predicting Numbers

When the output is a continuous value (number on a scale).

2. Classification: Predicting Categories

When the output is a discrete category (one of several options).

Binary vs Multi-Class Classification

Common Supervised Learning Algorithms

The Supervised Learning Workflow

Regression vs Classification: Quick Comparison

Key Takeaways

• Supervised learning uses labeled data (inputs + correct outputs)
• Regression predicts continuous values (numbers)
• Classification predicts discrete categories
• The workflow: collect data → split → train → evaluate → predict
• This is the most common type of ML in industry

Next, we'll explore unsupervised learning - where machines find patterns without labeled answers!