Data Science Data Science MCQ 3 — Questions and Answers
Question 1: What does a p-value of 0.03 indicate in hypothesis testing at a 0.05 significance level?
- Reject the null hypothesis (Correct answer)
- Accept the null hypothesis as proven true
- The result has no practical significance
- The sample size was too small
Correct answer: Reject the null hypothesis
Since 0.03 is below 0.05, the result is statistically significant and the null is rejected.
Question 2: Which algorithm builds an ensemble of trees sequentially, each correcting the errors of the previous?
- Gradient boosting (Correct answer)
- Random forest
- K-nearest neighbors
- Naive Bayes
Correct answer: Gradient boosting
Gradient boosting adds trees sequentially to minimize the residual errors of prior trees.
Question 3: What is the main advantage of using the F1 score over accuracy?
- It balances precision and recall, useful for imbalanced classes (Correct answer)
- It is always higher than accuracy
- It ignores false negatives
- It requires no ground truth labels
Correct 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.
Question 4: In a normal distribution, approximately what percentage of data falls within one standard deviation of the mean?
- 68% (Correct answer)
- 95%
- 50%
- 99.7%
Correct answer: 68%
By the empirical rule, about 68% of values lie within one standard deviation of the mean.
Question 5: Which preprocessing step scales features to a range typically between 0 and 1?
- Min-max normalization (Correct answer)
- One-hot encoding
- Label encoding
- Imputation with the mean
Correct answer: Min-max normalization
Min-max normalization rescales each feature to a fixed range, usually [0, 1].
Question 6: What problem does the ROC-AUC metric primarily evaluate?
- A classifier's ability to rank positives above negatives across thresholds (Correct answer)
- The training speed of a model
- The variance of the residuals
- The number of clusters in the data
Correct 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.
Question 7: Which technique addresses class imbalance by generating synthetic minority-class samples?
- SMOTE (Correct answer)
- PCA
- Dropout
- Batch normalization
Correct answer: SMOTE
SMOTE creates synthetic examples by interpolating between minority-class neighbors.
Data Science Data Science MCQ 3
What does a p-value of 0.03 indicate in hypothesis testing at a 0.05 significance level?