Classification — Google Cloud ML Engineer Practice Questions

Classification is a supervised learning task where a model predicts which discrete category an input belongs to, and it underpins many real-world ML applications such as fraud detection, image labeling, and spam filtering. The Google Cloud ML Engineer exam covers both binary and multi-class classification, including how to choose appropriate algorithms, loss functions, and output activation functions for each case. Candidates must understand how to evaluate classifiers using confusion matrices, ROC-AUC, precision, recall, and F1-score, and know how to tune decision thresholds to balance false positives and false negatives. BigQuery ML and Vertex AI AutoML both support classification tasks, so candidates should understand the trade-offs in choosing between them.

Free questions on classification

More classification questions in the full bank

• Which technique is commonly used to handle imbalanced datasets in classification? Unlock answer & explanation →
• Which evaluation metric is best for imbalanced classification problems? Unlock answer & explanation →
• When preparing data for a classification model, you discover that classes are imbalanced (95% negative, 5% positive). How should you handle this during training? Unlock answer & explanation →