Use the following classifiers (using the default configurations with the exception of the classifier of your own choice) and 10-fold Cross Validation:

i) IBk

ii) J48

iii) MultilayerPerceptron

, answer the following questions:

1) Which of the classifiers performed better in terms of the underrepresented class? explain your answer.

2) Consider the IBk classifier, given the default configuration why might this classifier be a poor fit for such an unbalanced classification problem?

3) In general terms suggest two approaches you might take to improve upon the classification of the underrepresented class. Discuss the advantages and disadvantages of each approach.