All terms

Classification

Predicting a discrete category or label for an input — spam vs not spam, language ID, sentiment class.

Machine Learning1 min read

Definition

Classification assigns each input to one of a set of classes. Binary classification has two labels; multiclass has many; multilabel allows several labels at once.

Metrics include accuracy, precision, recall, and F1 — especially important when classes are imbalanced.

In simple terms

Classification is sorting fruit into bins labeled apple, orange, or banana — each item gets a category tag.

Where you see it

  • Language identification for Somali vs English text.
  • Medical image disease vs healthy.
  • Intent classification in chatbots.

How it works

  1. 1.Label examples

    Each training item gets a class.

  2. 2.Train a model

    Learn decision boundaries or probabilities per class.

  3. 3.Predict

    Output the most likely class (or a probability distribution).

Why it matters

  • Classification is one of the most common production ML tasks across products and research.

Often confused

  • High accuracy always means a good classifier.

    On imbalanced data, a model that always predicts the majority class can score high accuracy while being useless.