1. After pre-processing a dataset, you are beginning an investigation of some data that has been classified. First, you created a model using ZeroR. The ZeroR model correctly classified 68.75% of the instances. As a next step, you run the J48 algorithm to build a classification tree. The box below shows some of the Weka output from the first run. (some stuff has been deleted from the Weka output to keep the question shorter. You dont need to know any of the deleted stuff.) What is your next step?

=== Run information ===

Scheme:weka.classifiers.trees.J48 -C 0.25 -M 2

Relation: CS560_FinalExamQ

{some stuff deleted}

Instances:192

Attributes:5

Test mode:10-fold cross-validation

=== Classifier model (full training set) ===

{some stuff deleted}

=== Stratified cross-validation ===

{some stuff deleted}

=== Summary ===

Correctly Classified Instances 129 67.1875 %

Incorrectly Classified Instances 63 32.8125 %

{some stuff deleted}

Total Number of Instances 192

=== Confusion Matrix ===

a b <-- classified as

128 4 | a = negative

59 1 | b = positive