In linear regression scoring, to ensure scoring data attributes fall within the training data ranges, which RapidMiner operator is used to align the ranges?

• A. Normalize
• B. Replace
• C. Sort
• D. Filter Examples

Answer: (D)

Keeping scoring data within what the model has seen during training is about filtering out any scoring records that fall outside the trained range. In RapidMiner, you use Filter Examples for this. It lets you keep only those scoring instances where each attribute lies between the minimum and maximum values observed in the training data. By applying a condition such as attribute >= training_min and attribute <= training_max (with training_min and training_max derived from the training set), you prevent out-of-range values from influencing the scoring, reducing extrapolation and keeping predictions more reliable. Normalization, for example, scales data but does not inherently restrict values to the training bounds, while other operators serve different purposes like filling in missing values or sorting.