WorksheetsThe Awakening Trial
Total questions: 25
Worksheet time: 13mins
1. The Sphinx awakens, demanding logic. Which measure is most affected by outliers?
Mean
Median
Mode
Range
2. The commander’s scroll holds ‘Male’, ‘Female’, ‘Other’. What type of data is this?
Age
Gender
Salary
Score
3. The Sphinx frowns at multicollinearity. What does it cause in regression?
Model underfitting
Biased coefficients
Unstable coefficients
Missing values
4. Two features mirror each other too closely in your model. What will you check?
p-value
VIF
RMSE
Z-score
5. The Sphinx reveals skewed income data. What transformation should you try first?
Square
Log
Standardize
Mean Substitution
6. The scroll has 10% missing entries in a skewed numerical column. Best imputation?
Mean
Median
Mode
Drop-down column
7. A chart in the temple compares parts of a whole. What kind is it?
Histogram
Box Plot
Pie Chart
Scatter Plot
8. The commander wants to track trends over time. Which visual should he use?
Pie Chart
Line Chart
Scatter Plot
Heat Map
9. What does a box in a box plot represent?
Standard Deviation
Confidence Interval
Mean and Median
Inter-quartile range
10. The ancient model classifies emails as spam or not. Which algorithm suits this best?
K-means
Linear Regression
Logistic Regression
PCA
11. The vaults echo with imbalance. The dataset has 80% of one class, 20% of another. What issue does this represent?
Outliers
Multicollinearity
Low Variance
Class Imbalance
12. The commander uses a regression model for prediction. R² = 0.05. What does this mean?
Excellent fit
Model explains very little
Model is overfitted
Data has no correlation
13. The ancient system records “Yes”/“No” as 1 and 0. What is this called?
Binning
Normalization
One-Hot Encoding
Binary Encoding
14. The Sphinx offers two features with high correlation. What risk is this in linear regression?
Heteroscedasticity
Multicollinearity
Overfitting
Low p-value
15. The cursed commander misused p-values. A very low p-value (< 0.05) indicates what?
Accept null hypothesis
Reject null hypothesis
Correlation is high
Data is skewed
16. The Realm of Visuals shows sales vs. time. Trendline moves upward.
Positive
Negative
Inverse
No correlation
17. A team wants to analyze a relationship between categorical variables. Choose:
t-test
ANOVA
Chi-square Test
Regression
18. A team builds a model that performs great on training but poorly on test data.
Overfitting
Underfitting
Low bias
High variance
19. In the Realm of Feedback, precision = 0.80, recall = 0.60. What is the F1 score?
0.68
0.70
0.60
0.75
20. The corrupted model misclassifies 30% of the time. What’s the accuracy?
70%
30%
85%
Not enough info
21. In the Realm of Excelor, what does =VLOOKUP() help with?
Vertical sorting
Fetching data based on key
Filtering duplicates
Creating pivot tables
22. The Sphinx hands over two variables: age and income. What chart shows their relationship?
Line Chart
Bar Chart
Scatter Plot
Pie Chart
23. The scroll shows missing values. Which is NOT a valid method to handle them?
Mean Imputation
Dropping rows
Creating dummy variable
Replacing with outlier
24. Which technique is used to reduce the number of features while preserving variance?
Clustering
PCA
Regression
Cross-validation
25. The Sphinx smiles—your trial nears completion. One last riddle before the first key is yours. To test two ad creatives on different user groups and determine which performs better, what method should you use?
Chi-square
A/B Testing
K-Means
PCA
