Skip to content

ACCA Strategic Professional · Advanced Performance Management · Data science and analytics

Zenith Retail plc wants to group its 2 million loyalty-card customers into segments with similar buying behaviour. No predefined segment labels exist, and management wants the algorithm to discover the groupings itself. Which machine learning approach is most appropriate?

Unsupervised learning using clustering is appropriate because the data has no predefined labels and the aim is to discover natural customer groupings. Supervised methods need labelled outcomes, and regression predicts numeric values rather than segments.

  1. ASupervised classification using labelled historical segments
  2. BUnsupervised learning using clusteringCorrect
  3. CLinear regression to forecast monthly sales
  4. DReinforcement learning with a reward for each purchase

Explanation

Clustering is an unsupervised technique that finds natural groupings in unlabelled data. Supervised classification needs labelled outcomes, which do not exist here. Regression predicts a numeric value rather than discovering groups.

Did you get it right without looking?

One question tells you little. A timed set on Data science and analytics shows your real accuracy, how long you take and where you lose marks.

Practise Data science and analyticsFree diagnostic mock

More Data science and analytics questions