In the logistic regression workflow, the ranges for all attributes in the scoring data must be within the ranges for the training data. Which operator can be used to match these ranges?

• A. Normalize
• B. Convert
• C. Filter Examples
• D. Cluster

Answer: (C)

Matching the ranges seen in training data to the scoring data is essential because the model expects feature values to lie within the same bounds it learned from. The way to enforce this in the workflow is to filter the scoring data so that every feature value falls between the minimum and maximum values observed in the training set. This keeps the scoring data within the model’s learned support and avoids making predictions on out-of-sample values that could lead to unreliable results. Normalization rescales values but can still produce inputs outside the training range, and converting or clustering don’t directly address enforcing the range constraint. Filtering is the right tool to ensure compatibility.