Skip to content

IAI Actuarial Core Principles · Risk Modelling and Survival Analysis · Elementary principles of machine learning

Which statement about the K-means algorithm is correct?

K-means can converge to a local optimum that depends on its initial centroids, so it is usually run from several starting points and the best result kept. It assigns points to the nearest centroid and needs no labels or dendrogram.

  1. AIt guarantees the globally optimal clustering regardless of the starting centroids
  2. BIts result can depend on the initial centroids, so it is often run from several starting pointsCorrect
  3. CIt requires a dendrogram to choose the number of clusters
  4. DIt assigns each point to the cluster with the farthest centroid
  5. It can only be used when class labels are known

Explanation

K-means iterates between assigning points to the nearest centroid and updating centroids, converging to a local minimum of within-cluster sum of squares. Different starts can give different solutions, so multiple random starts are used. The other statements describe incorrect features.

Did you get it right without looking?

One question tells you little. A timed set on Elementary principles of machine learning shows your real accuracy, how long you take and where you lose marks.

Practise Elementary principles of machine learningFree diagnostic mock

More Elementary principles of machine learning questions