Supervised vs Unsupervised Learning

Machine learning problems can be divided into two main categories:

• Supervised learning
• Unsupervised learning

The difference depends on whether we have labeled data.

---

Supervised Learning

In supervised learning, each training example includes:

• input features (X)
• a known target label (y)

The model learns a mapping from inputs to outputs.

In other words:

We know the correct answer during training.

---

Examples of Supervised Learning

• House price prediction (regression)
• Spam detection (classification)
• Disease diagnosis (classification)
• Energy consumption forecasting (regression)

Supervised learning problems are divided into: - Regression (continuous output) - Classification (discrete output)

---

What the Model Learns

The model tries to:

• minimize prediction error
• generalize to unseen data
• approximate the relationship between X and y

Supervised learning is goal-oriented: we explicitly define what we want to predict.

---

Unsupervised Learning

In unsupervised learning, the dataset contains:

• input features (X)
• no target labels

There is no “correct answer” provided.

The model tries to discover patterns, structure, or relationships within the data.

---

Examples of Unsupervised Learning

• Customer segmentation
• Anomaly detection
• Topic modeling
• Dimensionality reduction

---

What the Model Learns

The model may try to:

• group similar samples (clustering)
• reduce data complexity
• detect unusual patterns
• find hidden structure

Unsupervised learning is exploratory: we do not specify a prediction target.

---

Key Differences

Supervised Learning: Input (X) —> Model —> Known Target (y)

Unsupervised Learning: Input (X) —> Model —> Discovered Structure

---

When to Use Each Approach

Use supervised learning when: - labeled data is available - a clear prediction goal exists

Use unsupervised learning when: - labels are unavailable - the objective is exploration - you want to understand the structure of the data

---

Why This Distinction Matters

Understanding whether a problem is supervised or unsupervised:

• determines model choice
• determines evaluation strategy
• determines how performance is measured

It is one of the first decisions in any machine learning project.