wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

ANN Part 4 - Part 6 Quiz

Total questions: 20

Worksheet time: 10mins

Name
Class
Date
1.

Which of the following is not a way to generate dataset for ANN?

a)

Interview or survey

b)

Random guessing

c)

Hardware simulation

d)

Previous performance records

2.

In ANN datasets, what do the rows typically represent?

a)

Output classes

b)

Epochs

c)

Input variables

d)

Training accuracy

3.

For a dataset with 70% training and 30% testing, how many testing patterns are there in a 10-pattern dataset?

a)

7

b)

3

c)

5

d)

2

4.

What is the purpose of using a separate testing dataset in ANN?

a)

To reduce cost

b)

To check model generalisation

c)

To increase speed

d)

To avoid using historical data

5.

Which Python command is used to normalise input data in ANN?

a)

fit_transform()

b)

scale_input()

c)

sort_data()

d)

mean_adjust()

6.

What does denormalisation in ANN do?

a)

Removes noise from the data

b)

Reverts predicted values back to original scale

c)

Removes outliers

d)

Combines datasets

7.

What type of value does the output represent in a classification problem?

a)

Continuous

b)

Category

c)

Decimal

d)

Boolean only

8.

What are the two sequential phases in supervised ANN training?

a)

Backpropagation and testing

b)

Prediction and adjustment

c)

Forwarding and adjustment

d)

Training and saving

9.

What does backpropagation calculate in ANN training?

a)

Input size

b)

Gradients of error

c)

Memory usage

d)

Number of neurons

10.

Which of the following best describes the learning rate (α)?

a)

Number of hidden layers

b)

Rate of data input

c)

Size of weight adjustment during training

d)

Bias value update

11.

What happens if the learning rate α is too high?

a)

Model trains faster with perfect accuracy

b)

Model may overshoot optimal solution

c)

Model will stop learning

d)

Model will underfit

12.

What is the main role of the optimizer in ANN training?

a)

Compute gradients

b)

Choose number of layers

c)

Apply gradient updates to weights

d)

Set bias values

13.

What is the role of momentum (β) in ANN training?

a)

Avoid gradient vanishing

b)

Prevent overfitting

c)

Avoid local minima trap

d)

Increase learning rate

14.

Which is NOT a type of optimizer used in ANN?

a)

ADAM

b)

SGD

c)

ReLU

d)

RMSprop

15.

Which command in Keras sets the optimizer and loss function?

a)

model.add()

b)

model.fit()

c)

model.compile()

d)

model.train()

16.

What is the purpose of ANN testing?

a)

To increase training time

b)

To simulate unseen data

c)

To reset weights

d)

To normalise inputs

17.

What does the correlation coefficient (r) measure in ANN testing?

a)

Speed of convergence

b)

Difference between two datasets

c)

Relationship between predicted and actual output

d)

Number of output classes

18.

What is the desired value of correlation coefficient (r) for successful ANN testing?

a)

r < 0

b)

r = 0

c)

r ≥ 0.9

d)

r ≤ 0.5

19.

Which equation describes the regression line in ANN testing?

a)

Y = aX + b

b)

Y = mX + c

c)

Y = X + 1

d)

Y = wX + b

20.

What should you do if your correlation coefficient is below 0.9?

a)

Accept the result

b)

Reduce input variables

c)

Retrain the network

d)

Change the testing data