Choose Models for Imbalanced Data and Time-Series Forecasting

Q: Choose Models for Imbalanced Data and Time-Series Forecasting

This question evaluates competence in selecting and tuning models for time-series forecasting with trend and seasonality and for extremely imbalanced classification, while assessing understanding of OLS assumptions, trade-offs among tree ensemble methods, imbalance handling approaches, and end-to-end forecasting workflows.

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

Scenario

You must choose and tune models for (a) forecasting marketplace demand with seasonality and trend, and (b) detecting fraud where the positive class rate is only 0.2%.

Tasks

Ordinary Least Squares (OLS): Explain how OLS linear regression works and list its key assumptions.
Tree Ensembles: Compare gradient-boosted trees, random forests, and bagging. When would you prefer each?
Class Imbalance (0.2% positive): How would you handle this imbalance during model training and evaluation?
Time-Series Forecasting Workflow: Describe a full, practical workflow for modeling a series with trend and seasonality, including preprocessing, feature engineering, appropriate metrics, and time-aware cross-validation.

Hint

Address data preprocessing, feature engineering, resampling/weighting, proper metrics for imbalance, and cross-validation suited for temporal data.

Choose Models for Imbalanced Data and Time-Series Forecasting

Scenario

Tasks

Hint

Solution

Comments (0)