Load the dataset into a variable mydata. Remove any rows that contain missing values on any of the variables. Change the variable loan_status to binary (1 for Fully Paid and 0 for charged off). Change the variable home_ownership to a categorical variable (0 for RENT, 1 for MORTGAGE and 2 for OWN). Remove the word months in variable term and the symbol % in the variable int_rate. (2 Points) Split the dataset into 70% training and 30% testing using the sample() function. We are tying to predict loan_status so store the variable value in response.test and set the value to null on the test dataset. (1 Point) Run the logistic regression on trainData using glm() command with loan_amnt, funded_amnt, int_rate, term, annual_inc,dti and delinq_2yrs as input variables. Explain what the coefficient values for the variables int_rate and delinq_2yrs mean in plain English. (2 Points) Now use the predict() function on the test dataset to predict the probability of outcome. Set the predicted outcome to be 1 if probability is greater than 0.5 or else 0. Use the table() command to compare the predicted outcome and the actual. What are the values of precision, recall and overall accuracy? (2 Points) Now use the trainControl() and train() methods on mydata to perform 10-fold cross validation and use logistic regression. What are the values of precision, recall, overall accuracy and AUC? (Hint : You need to install the libraries caret and pROC before you run the models) (3 Points).