Introduction to Classification, the game
Spam or not? Which species? Loan or no loan? Learn how machines sort things into categories with five quick missions and a boss quiz.
Sort it into a box
Classification is supervised learning where the model learns from labelled examples and predicts a category for new data. Example: from a patient’s medical record, predict whether they have high blood sugar (1) or not (0).
Its sibling, regression, predicts a number instead of a category.
Three kinds of classification
- Binary: exactly two classes (spam / not spam).
- Multi-class: one of three or more classes (cat, dog or horse).
- Multi-label: each item can have several labels at once (a movie can be comedy and romance).
Classify by your neighbours
One of the simplest classifiers looks at the nearest labelled examples and takes a vote. Slide k (how many neighbours vote) and watch the predicted regions change for three iris species.
The playground uses R’s built-in iris data (150 flowers, 3 species) with just two features, sepal length and width, so you can see the decision regions. Every third flower is held out as a test set.
- ✓Try k = 1: every single flower gets its own territory+20 XP
- ✓Find the k with the best test accuracy+30 XP
Meet the algorithms
Classification is used everywhere: medical imaging, loan approval and credit scoring, predicting match results, spam and sentiment detection, sound recognition. Flip the cards to meet the main algorithm families (each has its own game lesson).
Defeat the Classification Boss 🐲
Six questions, three hearts.
You’re a Classification Champion!
Keep levelling up
You learned what classification is, how it differs from regression, the three types of classification problems, where it is used, and the main algorithm families.