The goal is to whether a new customer will accept a loan offer. This will serve as the basis for the design of a new campaign.

The file UniversalBank.xlsx contains data on $5000$ customers. The data include customer demographic information $($age$,$ income, etc.$),$ the customer$$s relationship with the bank $($mortgage$,$ securities account, etc.$),$ and the customer response to the last personal loan campaign $($Personal Loan$).$ Among these $5000$ customers, only $480(=9\mathrm{.}6\%)$ accepted the personal loan that was offered to them in the earlier campaign.

Partition the data into training $(60\%)$ and validation $(40\%)$ sets. Specify the success class as $1($loan acceptance$)$ and use the default cutoff value of $0\mathrm{.}5.$

$1)$ Using R perform a neutral net classification with all predictors except ID and ZIP code. Remember to transform categorical predictors with more than two categories into dummy variables first and scale numerical predictor variables to a $01.$

$2)$ Using R perform a deeper decision tree classification and prune with all predictors except ID and ZIP code.

$3)$ Based on the neutral net confusion matrix results versus the deeper decision tree confusion matrix results, identify which model method is best.

$4)$ Consider the following customer:

Age $=40,$ Experience $=10,$ Income $=84,$ Family $=2,$ CCAvg $=2,$ Education$\_1=0,$ Education$\_2=1,$ Education$\_3=0,$ Mortgage $=0,$ Securities Account $=0,$ CD Account $=0,$ Online $=1,$ and Credit Card $=1.$ How would this customer be classified using neutral net or decision tree based on which was identified as performing best?

$5)$ Provides R codes of Neutral Net Classification and Deeper Decision Tree Classification. Also please, follow the attached Rubric.

$6)$ Rubric

$**$ Accurately partitions the data into training and validation sets with a split of $60\%$ training and $40\%$ validation. The success class is correctly specified as $1($loan acceptance$),$ and the default cutoff value of $0\mathrm{.}5$ is used.

$**$ Neural net classification is performed accurately using all predictors except ID and ZIP code. Categorical predictors with more than two categories are appropriately transformed into dummy variables, and numerical predictors are scaled to a range of $01.$

$**$ Deeper decision tree classification is performed accurately using all predictors except ID and ZIP code. The tree is appropriately pruned, and all necessary predictor transformations are applied.

$**$ A clear comparison is made between the neural net and deeper decision tree models based on their confusion matrix results for both training and validation sets. The identification of the best$-$performing model is justified based on these results.

$**$ The customer is accurately classified using the best$-$performing model identified in part c$.$ A clear explanation is provided for how the classification was determined.