Which two modeling techniques handle both categorical and continuous target variables? (Choose two.)

Question

Options

• AQUEST
• BCHAID
• CC5.0
• DC & R Tree

• A
  13% (3)
• B
  79% (19)
• D
  8% (2)

Explanation

Note: Based on standard data mining definitions, the stated answer (B, C) appears to contain an error - the more accurate correct pair is B and D. Here's why:

CHAID (B) handles both target types: it uses chi-square tests for categorical targets and F-tests for continuous targets, making it genuinely dual-purpose. C & R Tree (D) is based on CART (Classification and Regression Trees) - the name itself signals support for both categorical (classification) and continuous (regression) targets.

C5.0 (C) is a classification-only algorithm descended from C4.5/ID3; its target field must be categorical, so it does not handle continuous targets. QUEST (A) is designed for binary classification only (categorical targets), so it also cannot handle continuous target variables.

Memory tip: Focus on algorithm names - C & R literally stands for "Classification and Regression," so it handles both. CHAID uses different statistical tests depending on target type, signaling flexibility. Any algorithm with "classification" exclusively in its lineage (C5.0, QUEST) is your distractor.

If this question came from an IBM SPSS Modeler study guide, double-check the source material - in that platform, B (CHAID) and D (C & R Tree) are the nodes documented as supporting both target types.

No community discussion yet for this question.