AI-900 Practice Question: Describe fundamental principles of machine learning on Azure
A data scientist trains a binary classification model to detect fraudulent transactions. The dataset contains only 1% fraudulent cases. The model predicts 'not fraudulent' for all transactions and achieves 99% accuracy. Which metric would best reveal the model's poor performance on fraud detection?
⚠ Common exam trap
The trap here is that candidates see 99% accuracy and assume the model is performing well, failing to recognize that accuracy is a poor metric for imbalanced datasets where the minority class (fraud) is the focus.
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
✓
Recall
Recall (sensitivity) measures the proportion of actual positive cases (fraudulent transactions) correctly identified by the model. With 1% fraud, a model that predicts 'not fraudulent' for all transactions will have a recall of 0% because it fails to catch any true positives, despite 99% accuracy. This makes recall the best metric to reveal the model's inability to detect fraud.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Precision
Why it's wrong here
Precision measures how many predicted frauds were actually fraud: TP / (TP + FP). Here the model never predicts fraud, so both TP and FP are zero, making precision undefined (0/0). If one arbitrarily defines it as 0, that only says the model made no positive claims, but it says nothing about the fraudulent transactions it completely missed. Recall is the direct measure of missed fraud, so precision is not the best diagnostic here.
- ✓
Recall
Why this is correct
Recall, also sensitivity or true positive rate, is computed as TP / (TP + FN). Because the model fails to flag a single fraudulent transaction, TP = 0 and all actual frauds become false negatives, yielding recall = 0. This zero score precisely exposes the model's total failure to catch fraud, which is the business-critical goal. In imbalanced fraud detection, recall is the primary metric because the cost of missing fraud is far higher than false alarms.
- ✗
F1 score
Why it's wrong here
The F1 score is the harmonic mean of precision and recall: 2 * (P * R) / (P + R). Since recall is 0, F1 also becomes 0, so it does indicate the model is failing overall. However, F1 collapses precision and recall into a single number, and if precision were undefined, F1 would be problematic or misleading. Recall alone clearly and directly attributes the failure to the model's inability to find any positives, so it is a better choice than F1 for diagnosing this exact problem.
- ✗
Accuracy
Why it's wrong here
Accuracy is the ratio of correct predictions to total predictions: (TP + TN) / (TP + TN + FP + FN). With 99% of transactions being legitimate, a model that always predicts 'not fraud' achieves 99% accuracy because it correctly classifies all true negatives. But it catches zero actual frauds, so the accuracy number hides a complete failure on the minority class. Accuracy is misleading for imbalanced datasets because it is dominated by the majority class and doesn't penalize missed positives.
Go deeper
Related to this question
Learn chapter
Regression and Classification
Key term
Model
In IT and AI, a model is a trained mathematical representation that learns patterns from data to make predictions or decisions.
Key term
Classification
Classification is a supervised machine learning technique used to predict a category or class label for new data based on patterns learned from labeled training data.
About these practice questions
One of 985 original AI-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.