Unit U4.3 · embed after: the classification section (k-NN)

K-Nearest Neighbours - region view, metrics and weighted voting

Data Science & Big Data Analytics • ITec4121 • Wolkite University • Online-First

Drop a query point on the canvas (click), choose k (1–9), a distance metric (Euclidean, Manhattan or Chebyshev) and a voting rule (majority or inverse-distance), pick a 24/36/48-point training set, and toggle the decision-region shading. Watch the nearest neighbours, distances and the vote - the same decisions the U4.3 classification exercise asks you to justify.

Playground

k = Metric = Vote =
Points = show decision regions

Click inside the canvas to place the unlabelled query point (black ●). Coloured points are labelled training examples (blue, green, red).