Equation Deck

Question 1

What is the equation for entropy?

Answer 1

Remember, entropy calculations are based on the data subsets created when analysing different variables with respect to the target variable.

Question 2

Formula for information gain

Answer 2

Question 3

Logistic Function

Answer 3

Question 4

Softmax Function

Answer 4

Question 5

How do you calculate the accuracy from the confusion matrix?

Answer 5

Acc = (TP + TN)/n

Where n is the total of the confusion matrix.

Question 6

How do you calculate the sensitivity from the confusion matrix?

Answer 6

Sensitivity = (TP)/(TP+FN)

Question 7

How do you calculate the specificity from the confusion matrix?

Answer 7

Specificity = TN/(TN+FP)

Essentially the right column.

Question 8

Expected benefit of targeting equation

Answer 8

Where v_R is the value we from a response and v_NR is the value we get from no response. Since everyone either responds or does not, our estimate of the probability of not responding is just (1-p_R(x))

where vR is the value we get from a response and vNR is the value we get from no response. Since everyone either responds or does not, our estimate of the probability of not re‐ sponding is just (1 - pR(x)).

Question 9

Equation for expected profit

Answer 9

b stands for benefit and c stands for cost.

Question 10

Write down the Bayes’ optimal classifier formula

Answer 10