Question 9
Domain 1You are developing a model to detect fraudulent credit card transactions. You need to prioritize detection, because missing even one fraudulent transaction could severely impact the credit card holder. You used AutoML to train a model on users' profile information and credit card transaction data. After training the initial model, you notice that the model is failing to detect many fraudulent transactions. How should you increase the number of fraudulent transactions that are detected?
Correct answer: D
Explanation
A lower probability threshold labels more transactions as fraudulent, which increases recall and catches more true positives. In binary classification, the threshold controls the tradeoff between precision and recall; lowering it is the standard way to reduce missed fraud when “missing even one fraudulent transaction” is costly.
Why each option is right or wrong
A. Add more non-fraudulent examples to the training set.
B. Reduce the maximum number of node hours for training.
C. Increase the probability threshold to classify a fraudulent transaction.
D. Decrease the probability threshold to classify a fraudulent transaction.
Vertex AI AutoML binary classification returns a probability score for the positive class, and the decision boundary is set by a threshold rather than fixed at 0.5. Lowering that threshold makes the classifier label more transactions as fraudulent, which increases recall and reduces false negatives—the right tradeoff when the business requirement is to miss as few fraud cases as possible.