Quiz title: C2  Making Predictions from Data


Title: Q1 Multiple Choice
Points: 1
1. What is a machine learning model fundamentally designed to do?
a) Store large amounts of data
b) Replace human decision-making entirely
*c) Learn patterns from examples to make predictions on new data
d) Memorize specific training examples
 
Title: Q2 Multiple Choice
Points: 1
2. According to the text, what percentage of Amazon's revenue comes from their recommendation system?
a) 25%
*b) 35%
c) 45%
d) 55%
 
Title: Q3 Multiple Choice
Points: 1
3. What is the key difference between regression and classification problems?
a) Regression is more accurate than classification
*b) Regression predicts specific numerical values, classification predicts categories
c) Classification is only used for binary problems
d) Regression requires more training data
 
Title: Q4 Multiple Choice
Points: 1
4. Which type of machine learning problem would predicting house prices be?
*a) Regression
b) Classification
c) Clustering
d) Association
 
Title: Q5 Multiple Choice
Points: 1
5. What does it mean when a model "overfits" to training data?
a) The model is too simple to capture patterns
*b) The model memorizes specific examples rather than learning general patterns
c) The model requires more training time
d) The model has too few parameters
 
Title: Q6 Multiple Choice
Points: 1
6. In email spam detection, what type of machine learning problem is this?
a) Regression
*b) Binary classification
c) Multi-class classification
d) Optimization
 
Title: Q7 Multiple Choice
Points: 1
7. What is the main advantage of linear regression?
a) It always provides the most accurate predictions
*b) It's interpretable and shows how each factor influences the prediction
c) It works well with any type of data
d) It requires no parameter tuning
 
Title: Q8 Multiple Choice
Points: 1
8. What is collaborative filtering in recommendation systems?
a) Filtering out inappropriate content
*b) Finding users with similar preferences and suggesting items they liked
c) Using content characteristics to make recommendations
d) Removing duplicate recommendations
 
Title: Q9 Multiple Choice
Points: 1
9. According to the text, how much does Netflix estimate their recommendation system saves them annually?
a) $500 million
*b) Over $1 billion
c) $2 billion
d) $750 million
 
Title: Q10 Multiple Choice
Points: 1
10. What is the difference between binary and multi-class classification?
*a) Binary chooses between two options, multi-class chooses among three or more
b) Binary is more accurate than multi-class
c) Multi-class is only used for text data
d) Binary requires less training data
 
Title: Q11 Multiple Choice
Points: 1
11. What are hyperparameters in machine learning?
a) The patterns learned from data
b) The input variables used for prediction
*c) Adjustable settings that control how algorithms operate
d) The output predictions of a model
 
Title: Q12 Multiple Choice
Points: 1
12. What is the main purpose of feature importance analysis?
a) To reduce the size of the dataset
*b) To understand which input variables contribute most to accurate predictions
c) To increase model accuracy
d) To speed up training time
 
Title: Q13 Multiple Choice
Points: 1
13. What is content-based filtering in recommendation systems?
a) Filtering out low-quality content
*b) Analyzing item characteristics to recommend similar items
c) Using user ratings to make recommendations
d) Removing biased recommendations
 
Title: Q14 Multiple Choice
Points: 1
14. What is the main goal of optimization in machine learning?
a) To make models run faster
*b) To find the best solution from many possible alternatives
c) To reduce the amount of training data needed
d) To eliminate all prediction errors
 
Title: Q15 Multiple Choice
Points: 1
15. What is dynamic pricing an example of?
a) Static optimization
*b) Real-time optimization
c) Batch processing
d) Data preprocessing
 
Title: Q16 Multiple Choice
Points: 1
16. What do decision trees for regression create?
a) Linear equations
*b) A series of yes/no questions that lead to predictions
c) Probability distributions
d) Neural network connections
 
Title: Q17 Multiple Choice
Points: 1
17. What is the key insight about machine learning models mentioned in the text?
a) They always provide perfect predictions
*b) They learn general rules that can be applied to new situations
c) They only work with numerical data
d) They require constant human supervision
 
Title: Q18 Multiple Choice
Points: 1
18. What is polynomial regression better suited for compared to linear regression?
a) Smaller datasets
*b) Relationships that are curved rather than straight
c) Binary classification problems
d) Text data analysis
 
Title: Q19 Multiple Choice
Points: 1
19. What is the main advantage of random forests over individual decision trees?
a) They are faster to train
*b) They combine many trees to make more robust predictions
c) They require less data
d) They are easier to interpret
 
Title: Q20 Multiple Choice
Points: 1
20. What is matrix factorization used for in recommendation systems?
a) Reducing computational complexity
*b) Identifying hidden factors that influence user preferences
c) Filtering out inappropriate content
d) Speeding up recommendation generation
 
Title: Q21 Multiple Choice
Points: 1
21. What is the main difference between greedy algorithms and other optimization approaches?
*a) They make the best choice at each step without considering future consequences
b) They always find the globally optimal solution
c) They require more computational resources
d) They only work with linear problems
 
Title: Q22 Multiple Choice
Points: 1
22. What are SHAP values used for?
a) Training machine learning models
*b) Understanding how much each input variable contributed to individual predictions
c) Optimizing model parameters
d) Reducing overfitting
 
Title: Q23 Multiple Choice
Points: 1
23. What is the main purpose of AutoML tools in model tuning?
a) To replace human data scientists entirely
*b) To automate the testing of different parameter combinations
c) To collect training data automatically
d) To deploy models to production
 
Title: Q24 Multiple Choice
Points: 1
24. According to the text, what is the most important consideration when choosing between regression and classification?
a) The size of your dataset
b) The computational resources available
*c) What type of answer you need - a specific number or a category
d) The accuracy requirements of your application
 
Title: Q25 Multiple Choice
Points: 1
25. What is the key message about successful predictive modeling?
a) Always use the most complex algorithms available
*b) The best model provides actionable insights and helps people make better decisions
c) Accuracy is the only metric that matters
d) Models should replace human judgment entirely
 