The following evaluation indicators belonging to the regression algorithm are?

Question

Options

• ARecall rate
• BConfusion matrix
• CMean square error
• DAccuracy

• A
  3% (1)
• B
  12% (4)
• C
  76% (25)
• D
  9% (3)

Explanation

Mean Square Error (MSE) is the correct evaluation metric for regression because regression algorithms predict continuous numerical values, and MSE measures the average squared difference between predicted and actual values - a direct measure of prediction error magnitude.

• A (Recall rate) and D (Accuracy) are classification metrics. They measure how well a model categorizes discrete labels (e.g., "spam" vs "not spam"), which is meaningless for continuous outputs like predicting house prices.
• B (Confusion matrix) is also a classification tool - it tabulates true positives, false positives, etc., which only apply when outputs are discrete classes, not continuous numbers.

Memory tip: Think "Regression → Real numbers → MSE (and its relatives like MAE, RMSE)." If you can plot predicted vs. actual on a number line and care about distance, you're in regression territory. If you're sorting things into buckets, you're in classification territory - and that's where accuracy, recall, and confusion matrices live.

No community discussion yet for this question.