AI0-001 AI Models and Data Engineering Practice Question
A machine learning team is deploying a model that predicts loan default probabilities. The model outputs a probability score, and the team wants to convert it into a binary decision (default/no default). The costs of false positives and false negatives are not equal; a false negative (predicting no default when the customer defaults) is five times more costly than a false positive. Which approach best optimizes the decision threshold?
⚠ Common exam trap
The trap here is defaulting to 0.5 or accuracy-based thresholds, ignoring that the cost of a false negative is five times that of a false positive.
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
✓
Choose a threshold that minimizes the expected cost, weighting false negatives five times more than false positives.
The optimal decision threshold minimizes expected cost, which requires weighting false negatives according to their higher cost. Lowering the threshold increases the number of predicted defaults, reducing expensive false negatives at the expense of more false positives. This cost-sensitive approach aligns model decisions with business objectives, unlike accuracy maximization or arbitrary thresholds.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Set the threshold to 0.5 to balance false positives and false negatives equally.
Why it's wrong here
A threshold of 0.5 assumes equal costs for false positives and false negatives. Since a false negative is five times more costly, this threshold would lead to too many costly false negatives. It does not account for the asymmetric cost structure and would result in suboptimal business outcomes.
- ✗
Use the threshold that maximizes overall accuracy on the validation set.
Why it's wrong here
Maximizing accuracy assumes equal costs and balanced classes. With imbalanced costs and likely imbalanced classes, accuracy is not a reliable metric. A threshold that maximizes accuracy may still produce many costly false negatives. It ignores the business cost asymmetry and can lead to poor decisions.
- ✓
Choose a threshold that minimizes the expected cost, weighting false negatives five times more than false positives.
Why this is correct
The optimal threshold minimizes expected cost given the cost matrix. By weighting false negatives five times more, the threshold will be lowered to predict more defaults, reducing costly false negatives. This approach directly incorporates business costs and yields the most cost-effective decisions.
- ✗
Set the threshold to the prevalence of defaults in the training data.
Why it's wrong here
Using the prevalence as the threshold is a heuristic that does not consider costs. It may be appropriate if the goal is to match the base rate, but it does not minimize expected cost. With asymmetric costs, this threshold is unlikely to be optimal and could result in excessive false negatives.
About these practice questions
One of 962 original AI0-001 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 and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official CompTIA exam blueprint
This AI0-001 practice question is part of Courseiva's free CompTIA 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 AI0-001 exam.