Free Printable ML#Module#3and#4

1.

Which of the following statement about ANN is false?

a)

ANN is preferred when Input is high-dimensional discrete or real-valued

b)

ANN is preferred when less training time is desired

c)

ANN is preferred when input data contain noisy data

d)

ANN is preferred when Output is discrete or real valued

2.

In the perceptron training rule, weights are updated as follows.
$W_i\ =\ W_i+\ \Delta W_{i\ }$ Where $\Delta W_{i\ }\ =\$ -----------?

a)

$\eta\left(t\ -o\right)x_{i\ }$

b)

$\left(t\ -\ o\right)x_i$

c)

$\eta\left(t\ -\ o\right)$

d)

$\eta\left(t-o\right)w_i$

3.

What is the derivative of the sigmoid function $\sigma\left(x\right)\ =\ \frac{1}{\left(1\ +\ e^{-x}\right)}$ ?

a)

$\sigma\left(x\right)\left(1\ -\ \sigma\left(x\right)\right)$

b)

$\sigma\left(x\right)\left(1-\sigma\left(x\right)^2\right)$

c)

$\left(1\ -\sigma\left(x\right)\right)$

d)

$\frac{\sigma\left(x\right)}{\left(1-\sigma\left(x\right)\right)}$

4.

Consider a decision tree classifier. Suppose there are 9 instances and an attribute is randomly selected and that attribute split the examples into two classes A and B. class A contain 0 instance and class B contain 9 instance . Then which of the following inference is more appropriate?

a)

Entropy is 1.

b)

Entropy is high

c)

Entropy is low

d)

Entropy is 0

5.

GainRatio(S,A) = ----------------------------------

a)

Gain(S,A)/Split-information(S,A)

b)

Gain(S,A) * Split-information(S,A)

c)

Gain (S, A) * Entropy(S)

d)

None

6.

Which of the following statement about Naive Bayes classifier is false?

a)

Can successfully train on small data set

b)

Good for text classification, good for multiclass classification

c)

Quick and simple calculation since it is naive

d)

Can learn the relationship among the features because assumes feature independence

ML#Module#3and#4

Total questions: 6

Worksheet time: 6mins

Similar Resources on Wayground