Studyledge

Artificial Intelligence: Question 1

Syllabus 18.1

Multiple choice A2 1 mark

A video streaming service holds a record for each of 5,000 past subscribers. Each record already states the genres that subscriber watched and a label showing whether that subscriber went on to cancel their subscription or not. The company wants to train a machine learning model on this labelled data so that it can predict whether a new subscriber is likely to cancel.

Which category of machine learning is being described?

Choose an answer to check it, then compare with the worked solution below.
Show worked solution Hide worked solution

Worked solution

Identifying the category of machine learning

The company already knows, for every one of its 5,000 past subscribers, both:

  • the input (the genres watched), and
  • the correct output (whether that subscriber cancelled or not). This is the label.

Training a model on data where each example already carries its correct, known output is exactly what supervised learning means. The model uses these labelled examples to learn the relationship between the input and the output, so that it can then predict the (unknown) label. Cancel or not. For a brand new subscriber it has never seen before.

Why the other options are wrong

  • A (unsupervised learning): unsupervised learning trains on data that has no known correct output attached. The model is left to find patterns or groupings in the data by itself. Here, every past record already comes with its outcome labelled, so this is not the case.
  • C (reinforcement learning): reinforcement learning involves an agent taking actions in an environment and learning from reward or penalty signals it receives as a consequence of those actions, through trial and error. There is no agent taking actions here, and no reward signal, just a fixed set of already-labelled historical records.
  • D (deep learning): deep learning describes artificial neural networks built with many hidden layers. The size of the dataset alone does not make a task “deep learning”. Nothing in the scenario states that a many-layered neural network is being used.

Final answer

B. This is supervised learning, because the model is trained on labelled data in which the correct output for every example is already known.