Data Science MCQ Flashcards

What does a p-value of 0.03 indicate in hypothesis testing at a 0.05 significance level?

Answer: Reject the null hypothesis

Since 0.03 is below 0.05, the result is statistically significant and the null is rejected.

Which algorithm builds an ensemble of trees sequentially, each correcting the errors of the previous?

Answer: Gradient boosting

Gradient boosting adds trees sequentially to minimize the residual errors of prior trees.

What is the main advantage of using the F1 score over accuracy?

Answer: It balances precision and recall, useful for imbalanced classes

The F1 score is the harmonic mean of precision and recall, helpful when classes are imbalanced.

In a normal distribution, approximately what percentage of data falls within one standard deviation of the mean?

Answer: 68%

By the empirical rule, about 68% of values lie within one standard deviation of the mean.

Which preprocessing step scales features to a range typically between 0 and 1?

Answer: Min-max normalization

Min-max normalization rescales each feature to a fixed range, usually [0, 1].

What problem does the ROC-AUC metric primarily evaluate?

Answer: A classifier's ability to rank positives above negatives across thresholds

ROC-AUC measures how well the model separates classes across all decision thresholds.

Which technique addresses class imbalance by generating synthetic minority-class samples?

Answer: SMOTE

SMOTE creates synthetic examples by interpolating between minority-class neighbors.