Build Classifier: Evaluate with AUROC for Imbalanced Data

Q: Build Classifier: Evaluate with AUROC for Imbalanced Data

This question evaluates a data scientist's competency in building and evaluating binary classifiers for imbalanced datasets, focusing on model design, evaluation metric selection, and performance interpretation.

Q: How do I approach Machine Learning interview questions?

Machine Learning questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master machine learning interviews.

Question

Detecting Dead Links: Build and Evaluate a Classifier

Scenario

You have a dataset of 1,000 URLs labeled as good (alive) or bad (dead). The classes are likely imbalanced (e.g., far fewer dead links than good ones).

Task

Describe how you would build the classifier end-to-end (data prep, features, model, validation, and deployment considerations).
Explain which evaluation metric(s) you would choose for imbalanced data.
Clarify why AUROC might be preferred over accuracy when the classes are imbalanced.

Hint: A strong baseline is logistic regression with class-imbalance-aware metrics.

Build Classifier: Evaluate with AUROC for Imbalanced Data

Detecting Dead Links: Build and Evaluate a Classifier

Scenario

Task

Solution

Comments (0)