Compare the results from Exercise 13 with the results from the EDA and decision tree analysis in Chapters 3 and 6. Discuss similarities and differences. Which analysis format do you prefer? Do you find a confluence of results?
Exercise 13
Set the minimum antecedent support to 1%, the minimum rule confidence to 5%, and the maximum number of antecedents to 1. Use rule confidence as your evaluation measure.
a. Find the association rule with the greatest lift.
b. Report the following for the rule in (a).
(i) Number of instances
(ii) Support % (as defined in this chapter)
(iii) Confidence %
(iv) Rule support %
(v) Lift
(vi) Deployability
c. Using hand calculations, show how the measures were calculated.
d. Explain, in terms of this data, what each of the measures in (c) means (you can skip (i)).