DMC Troubleshooting & Problem Resolution 5

Question 1

Stakeholders do not trust a high-performing black-box model's predictions. What is the best resolution step?

Accepted Answer

Provide interpretability such as feature importance or a simpler surrogate model

Answer

Explainability tools build trust and help validate that the model uses sensible signals.

Question 2

A scheduled scoring job runs out of memory on a large dataset. Which fix is most appropriate?

Accepted Answer

Process the data in batches or chunks

Answer

Batch processing limits memory use without changing the results.

Question 3

Sampling for a churn model produced a training set that does not represent the full customer base. What is this problem called?

Accepted Answer

Sampling bias

Answer

A non-representative sample biases the model and harms generalization.

Question 4

When diagnosing any failed data mining project, which step should come FIRST?

Accepted Answer

Reproduce and isolate the problem, checking data and process logs

Answer

Systematic troubleshooting begins by reproducing and isolating the issue before changing anything.

Question 5

A regression model's residuals fan out as predicted values increase. Which assumption is violated?

Accepted Answer

Homoscedasticity (constant variance)

Answer

Fanning residuals indicate heteroscedasticity, often fixed by transforming the target.

DMC - Data Mining Certification Practice Test