This assignment will involve evaluating the benefits of using meta learners on our dataset.

Using the same methods as in Part $4$ of Assignment $1($including $10-$fold cross$-$validation$),$ determine the optimal cost ratio for:

Cost sensitive classifier combined with bagging and J$48$

Cost sensitive classifier combined with bagging and Decision Stump

Cost sensitive classifier combined with boosting $($AdaBoostM$1)$ and J$48$

Cost sensitive classifier combined with boosting $($AdaBoostM$1)$ and Decision Stump

Use the default settings for the meta learners $($bagging and boosting$)$ and the learners $($J$48$ and Decision Stump$).$

How do the results of each classifier compare to each other, and to the cost$-$sensitive tree obtained in Part $4$ of Assignment $1?$ Analyze and explain.

Repeat all experiments with the number of iterations for each meta learner set to $25$ instead of $10.$ Compare these results to each other, and to your previous results from the $10-$iteration models. Again, analyze and explain.