COMM 414 Final questions with verified answers
A+ updated
1. regplot() creates a scatterplot and draws a linear regression line
together with the confidence interval at the same time
2. CART classification and regression trees
3. regression tree response variable continuous, classify to value ranges
4. classification trees response variable discrete, classify to value categories
5. overfitting Overfitting happens when a model describes a random error or
noise instead
of underlying relationship
6. What is a decreasing tree depth
common method
to prevent
overfitting in
deci- sion trees?
7. prune method to overcome overfitting
8. two types of user based and item based
collab- orative
filtering
9. user-based collabo- Find similar users to me and recommend what they like
rative filtering
10. item-based collabo- Find similar items to those that I have previously liked
rative filtering
11. monte carlo simula- A broad class of computational algorithms that rely on repeated
random sam-
tions pling to obtain numerical results
12. collaborative A method of making automatic predictions (filtering) about the interests of
filter- ing a user by collecting preferences or taste information from many
1/
3
A+ updated
1. regplot() creates a scatterplot and draws a linear regression line
together with the confidence interval at the same time
2. CART classification and regression trees
3. regression tree response variable continuous, classify to value ranges
4. classification trees response variable discrete, classify to value categories
5. overfitting Overfitting happens when a model describes a random error or
noise instead
of underlying relationship
6. What is a decreasing tree depth
common method
to prevent
overfitting in
deci- sion trees?
7. prune method to overcome overfitting
8. two types of user based and item based
collab- orative
filtering
9. user-based collabo- Find similar users to me and recommend what they like
rative filtering
10. item-based collabo- Find similar items to those that I have previously liked
rative filtering
11. monte carlo simula- A broad class of computational algorithms that rely on repeated
random sam-
tions pling to obtain numerical results
12. collaborative A method of making automatic predictions (filtering) about the interests of
filter- ing a user by collecting preferences or taste information from many
1/
3
