AIF-C01 Fundamentals of AI and ML Practice Question

A fintech startup is preparing its first machine learning project to detect fraudulent card transactions. The team must decide which characteristics make a problem well suited to supervised learning. Which TWO characteristics indicate that supervised learning is appropriate? (Choose two.)

⚠ Common exam trap

The trap here is assuming that a large transaction volume or a desire for explainability alone makes a problem supervised, when labeled outcomes and a defined target are what actually matter.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Historical transactions exist that were confirmed as fraudulent or legitimate by investigators.

Supervised learning needs labeled examples and a defined target to predict. Confirmed fraudulent or legitimate transactions supply those labels, and a discrete fraud-or-not decision supplies a suitable classification target. Unlabeled data, label-free exploration goals, and explainability preferences do not establish the labeled input-output pairing that supervised learning fundamentally depends on.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
The team wants the algorithm to discover hidden structure without any predefined output.
Why it's wrong here
Discovering hidden structure without predefined outputs describes unsupervised learning, not supervised learning. Supervised models are explicitly trained to reproduce a known target, so a goal of label-free exploration contradicts the approach. This characteristic would justify clustering or anomaly detection techniques rather than a supervised fraud classifier trained on confirmed cases.
✓
Historical transactions exist that were confirmed as fraudulent or legitimate by investigators.
Why this is correct
Confirmed historical outcomes are exactly the labeled data supervised learning requires. Each transaction carries a target the model can learn from, allowing it to map transaction features to a fraud or legitimate decision. Without such labels, a supervised classifier could not be trained, so this characteristic is a defining indicator that the approach fits.
✗
The team prefers a model that explains each prediction with feature importance values.
Why it's wrong here
Explainability is valuable for regulated fraud systems, but it is a model selection or governance preference rather than a defining condition for supervised learning. Both supervised and unsupervised techniques can be made interpretable to varying degrees. Wanting explanations does not itself require labels or a discrete target, so it does not indicate that supervised learning is the right paradigm.
✓
The goal is to predict a discrete outcome for each new transaction.
Why this is correct
Supervised classification predicts discrete labels, and fraud detection asks whether each transaction is fraudulent or not, a binary outcome. Combined with the availability of labeled history, a clear discrete target confirms that supervised learning is the appropriate framing. Continuous or purely descriptive goals would instead point toward regression or unsupervised methods.
✗
The dataset contains millions of unlabeled transactions with no investigator feedback.
Why it's wrong here
Unlabeled data with no confirmed outcomes cannot train a supervised model because there is no target to learn from. This characteristic actually points toward unsupervised anomaly detection instead. Although large volumes of transactions are useful, the absence of investigator feedback removes the essential ingredient of supervised learning, so it does not indicate that supervised learning is appropriate.

Go deeper

AIF-C01 Fundamentals of AI and ML Practice Question

Why each option matters

Option-by-option breakdown

Related to this question