Use the “CarsARM2.csv” data set to answer the following questions. Load the data set in Weka and construct a decision tree (J48) with the default parameters and the “Use training set” test option.
a. What is the overall prediction accuracy for this decision tree model?

b. What are the numbers of correctly and incorrectly classified instances?

c. What is the size of the tree and the number of leaves?

d. Modify the parameters to have minNumObj=25. What is the new accuracy? Comment on the

accuracy change as compared to the previous scenario.

e. What are the new numbers of correctly and incorrectly classified instances?

f. Interpret the confusion matrix.

g. Identify those attributes (if any) that do not affect a car model’s acceptable rating according

to this decision tree.

h. Estimate the acceptable rating for the following car models based on this decision tree:

Buy Maint Doors People Lugg Safety Accept

High Vhigh 2 2 Big High ?

Med Med 4 More Med High ?

Vhigh Low 5more 4 Big Low ?