Minority Class — Google Cloud ML Engineer Practice Questions

The minority class is the underrepresented category in an imbalanced classification dataset, and correctly identifying instances of it is often the primary business objective, such as detecting fraudulent transactions or rare medical conditions. The Google Cloud ML Engineer exam tests how model behavior degrades when the minority class has too few examples, and what techniques can restore adequate sensitivity. Candidates must understand the role of class weights, threshold adjustment, and resampling in improving minority-class recall without sacrificing overall precision. Monitoring minority-class performance after deployment, using Vertex AI Model Monitoring for skew and drift detection, is also relevant for production ML systems on GCP.

Free questions on minority class