Need to build data classifiers on the following datasets using both decision tree and the ensemble method, and evaluate the quality of the classifiers.

Split dataset into a training data set and a testing data set. Use the training set to build a classifier and test its performance on the test set. When build your classifiers, be careful about which is an attribute and which is the class label.

Include your R scripts and contingency table of each classifier you build in your report. For overall performance of each classifier, furthermore, report accuracy (sum on the diagonal of the contingency table / total number of test records). For each class, report accuracy, specificity, precision and recall. For each data set, you should give and explain your conclusion on whether the ensemble method improves the performance of the decision tree method.

Following is the data set

Nursery

URL: http://archive.ics.uci.edu/ml/datasets/Nursery

Number of attributes: 8

Number of records: 12960

Suggested split for tree classifier: 8000 for training and the rest for testing