ExamPark
Course menuAIF-C01

Exams / AIF-C01 / Use cases and ML techniques / Question 5

AIF-C01 practice question: Use cases and ML techniques (5 of 5)

Fundamentals of AI and ML A free question from our original AIF-C01 practice exam. Try it, then open the answer to see why each option is right or wrong.

A subscription business wants to predict which customers will cancel next month. It has five years of structured account data in a table with a column showing who cancelled. A developer proposes sending every customer record to a large foundation model with a prompt asking whether they will cancel. What is the MOST appropriate approach?

Choose one answer.

  1. A. Train a traditional supervised classification model on the labelled table
  2. B. Send each record to the foundation model, because larger models are always more accurate
  3. C. Use a diffusion model, because it can generate likely future customer behaviour
  4. D. Use unsupervised clustering, because the cancellation column is not needed
Show the answer and explanation

Answer: A

  • A. Right: Correct. Labelled tabular data and a yes or no outcome is a classic supervised classification problem, and a dedicated model is cheaper and easier to evaluate.
  • B. Wrong: Foundation models are general purpose and billed per token. For a narrow tabular prediction they add cost and are harder to validate and explain.
  • C. Wrong: Diffusion models generate content such as images. They are not used to predict a label from tabular data.
  • D. Wrong: The cancellation column is a label the model can learn from. Ignoring it throws away the most useful information.

Explanation

Foundation models are powerful for open-ended language and content tasks, but they are not automatically the best choice. For a well-defined prediction on labelled structured data, traditional ML is usually cheaper, faster, easier to measure with metrics such as precision and recall, and easier to explain to regulators.

Study

Keep practising

Answer questions with a score at the end, or practise just this topic with the explanation after each one. No sign-up needed.

Take the free AIF-C01 practice examPractise Use cases and ML techniquesAll Use cases and ML techniques questions