QUESTION THREE

3.1 A confusion matrix, such as one shown below, is used to evaluate the

performances of algorithms in machine learning.

ACTUAL

YES NO

28 4

NO 6 24

(30 MARKS)

PREDICTED YES

a. In your own words, explain what a confusion matrix is.

(4)

b. In the confusion matrix above, the values in the diagonal are high (28 and 24),

what does that tell you about the accuracy of the model that has this

confusion matrix?

(4)

c. What do high values for False Positive and False negative mean? Explain (4)

d. Explain one example in which a high number of False Positives would have

serious consequences.

(4)

e. Explain one example in which a high number of False Negatives would have

serious consequences.

(4)

3.2 Given a simple artificial neural network that does not have a hidden layer,

compute the output value given the following information:

There are 3 inputs(x1 = 30, x2 = 18, x3 = -40) and the value of the bias (b) is