Reference no: EM133448674

Question

Import the bank.csv file into Rstudio, Convert the data into tibble and convert all character variables into factor Split the dataset into test data(20%) and train data (80%) keeping the same ratio as for "y" column Use the train data and KNN classifier for predicting whether the customer will subscribe (the target class is Y column) Use test data to test the created model and get the performance of the model Use k-fold cross validation with k=20 repeated 50 times Create the confusion matrix for the created model Calculate accuracy and mmce for this model Does the cross validation improved the performance(compare performances from parts e and h, of yes by how much?