Skip to content

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

Wexford Foods has a large set of customer purchase records with no pre-defined labels. It wants an algorithm to discover natural groups of customers with similar buying behaviour to design targeted promotions. Which approach is most appropriate?

Unsupervised learning, such as clustering, is most appropriate. The data has no pre-defined labels and the aim is to discover natural customer groupings. Supervised classification and regression require a known target outcome, and manual invoice sampling does not identify behavioural segments.

  1. ASupervised classification using labelled outcomes
  2. BUnsupervised learning, such as clusteringCorrect
  3. CLinear regression to predict a single sales value
  4. DManual sampling of invoices for audit testing

Explanation

With no labelled outcomes and a goal of finding hidden groupings, unsupervised learning such as clustering is suitable. Supervised classification and regression need a known target variable to learn from. Manual invoice sampling is an audit technique and does not segment customers.

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