What Is Classification?

What Is Classification?

Classification is used when we want a machine learning model to decide which category an input belongs to.

The model is not trying to predict an exact numerical value. It is trying to choose a category.

What Is a Class?

A class is simply a category or group that an input can belong to.

Here, Spam, Not Spam, Pass, and Fail are examples of classes.

Simple Example — Pass or Fail

Suppose we want to predict whether a student will pass an exam based on the number of hours they studied.

The model learns patterns from these examples.

Classification Uses Features and Labels

Classification uses the same basic idea of features and labels that we learned earlier.

For classification, the label represents a category such as Pass, Fail, Spam, or Not Spam.

Classification vs Regression

This is one of the most important differences to understand.

Real-World Example — Spam Detection

Classification is commonly used to identify whether an email is spam.

The model can use information such as:

The possible classes in this example are:

Real-World Example — Fraud Detection

A bank can use classification to determine whether a transaction is potentially fraudulent.

The Model Learns From Examples

A classification model learns from training examples where the correct class is already known.

Classification Can Use Probabilities

A classification model may internally estimate how likely each class is.

Based on these probabilities, the model may choose Pass as the predicted class.

Check Your Understanding