A Machine Learning Specialist trained a regression model, but the first iteration needs optimizing. The Specialist needs to understand whether the model is more frequently overestimating or…

The correct answer is B. Residual plots. Residual plots graph the residuals (residual = actual value − predicted value) for each data point. If the model overestimates, predicted values are consistently higher than actuals, so residuals are predominantly negative. If it underestimates, residuals are predominantly…

Modeling

Question

A Machine Learning Specialist trained a regression model, but the first iteration needs optimizing. The Specialist needs to understand whether the model is more frequently overestimating or underestimating the target. What option can the Specialist use to determine whether it is overestimating or underestimating the target value?

Options

ARoot Mean Square Error (RMSE)
BResidual plots
CArea under the curve
DConfusion matrix

How the community answered

(47 responses)

A
2% (1)
B
91% (43)
C
2% (1)
D
4% (2)

Explanation

Residual plots graph the residuals (residual = actual value − predicted value) for each data point. If the model overestimates, predicted values are consistently higher than actuals, so residuals are predominantly negative. If it underestimates, residuals are predominantly positive. A residual plot makes this directional bias immediately visible as a systematic skew above or below the zero line. RMSE (Root Mean Square Error) measures the magnitude of errors but squares them, losing sign information - it cannot distinguish over- from underestimation. Area under the curve (AUC) is a classification metric, not applicable to regression. A confusion matrix is also a classification tool that has no meaning for continuous regression output.

Topics

#Regression model evaluation#Residual analysis#Model diagnostics#Bias detection

Community Discussion

No community discussion yet for this question.

Full MLS-C01 Practice