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.Label examples
Each training item gets a class.
2.Train a model
Learn decision boundaries or probabilities per class.
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.