wayground logo

Free Printable Worksheets

Font size

S
M
L
XL
Worksheets

Titanic Challenge

Total questions: 16

Worksheet time: 48mins

Name
Class
Date
1.

For the Titanic dataset, what is the split ratio for training and test subsets?

a)

Approximately 20% training and 80% test.

b)

Approximately 68% training and 32% test.

c)

Approximately 80% training and 20% test.

d)

Approximately 70% training and 30% test.

2.

Which features have missing values?

4 lines
3.

How many new features are obtained after processing and transforming the original Cabin variable?

(a)  

4.

Which passengers have a higher chance of survival?

a)

Male passengers.

b)

First class passengers.

c)

Passengers that boarded the ship in Cherbourg.

d)

Passengers that boarded the ship in Queenstown.

5.

The Age column in the dataset has missing values. These missing values can be handeled using the following methods.

a)

Imputed with the median.

b)

Imputed with the mean.

c)

Predicted with a Decision Tree.

d)

Dropped from the dataset.

6.

What are the two most important features for predicting the missing values in Age?

a)

Rare and Miss.

b)

IsMale and Rare.

c)

Pclass and IsMale.

d)

Pclass and Parch.

7.

Which port of embarkation had the highest number of passengers?

a)

New York.

b)

Southampton.

c)

Queenstown.

d)

Cherbourg.

8.

What features could be one-hot encoded without drastically increasing the dimensionality of the data?

4 lines
9.

How could the missing data in the Embarked feature be handled?

a)

Imputed with the median.

b)

Imputed with the mean.

c)

Imputed with the mode.

d)

Dropped from the dataset.

10.

What percentage of passengers from the training data survived the Titanic disaster?

a)

About 38%.

b)

About 50%.

c)

About 68%.

d)

About 20%.

11.

If we use cross-validation with 10 folds, what is the number of samples in the validation subset for each cross-validation round?

a)

Approximately 42.

b)

Approximately 89.

c)

Approximately 890.

d)

Approximately 420.

12.

In the jupyter notebook, what new features are created by combining the SibSp and Parch features?

a)

InGroup

b)

InWcg

c)

WcgAllSurvived

d)

WcgAllDied

13.

Which method is used to perform hyperparameter tuning in the jupyter notebook?

a)

GridSearchCV

b)

RandomizedSearchCV

c)

Manual search

d)

Bayesian Optimization

14.

Which of the following features is not used in the final model training after data preparation?

a)

Survived

b)

AgeBand

c)

Pclass

d)

Name

15.

How many unique combinations of parameters with RandomizedSearchCV try?

a)

200

b)

100

c)

20

d)

10

16.

The hyperparameters we search for K-nearest Neighbors are (a)   ?