Identify and Fix Predictive Model Performance Gaps

Q: Identify and Fix Predictive Model Performance Gaps

This question evaluates a data scientist's competency in feature engineering (especially temporal encoding), feature standardization, and handling imbalanced datasets for predictive model performance within the Machine Learning domain.

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

Model Review: Month Encoding, Feature Scaling, and Imbalanced Data

Context

You are auditing an existing predictive model for operational performance. The current implementation encodes calendar month as a continuous variable and is trained on imbalanced data. Address the following:

Questions

Month as a continuous variable
- What problems can arise if the model treats calendar month (e.g., Jan=1, ..., Dec=12) as a continuous feature?
- How would you fix this encoding to capture seasonality correctly?
Feature standardization
- Why is standardizing predictors important before fitting certain models?
- What can go wrong if you skip standardization?
Imbalanced data and recall
- Your training data are highly imbalanced. Describe two concrete ways to adjust the loss function or the evaluation/thresholding so recall is properly rewarded.

Identify and Fix Predictive Model Performance Gaps

Model Review: Month Encoding, Feature Scaling, and Imbalanced Data

Context

Questions

Solution

Comments (0)