In this problem, you are going to build predictive models for bank telemarketing problem. The data is related with direct marketing campaigns of a Portuguese banking institution. The marketing campaigns were based on phone calls. Often, more than one contact to the same client was required, in order to access if the product $($bank term deposit$)$ would be $(\text{'}$yes$\text{'})$ or not $(\text{'}$no$\text{'})$ subscribed. You are given bank$-$ additional$-$full.csv file containing the data set.

Here is the information on the data attributes:

Input variables:

# bank client data:

$1-$ age $($numeric$)$

$2-$ job : type of job $($categorical: 'admin.$\text{'},$'blue$-$collar','entrepreneur','housemaid','management','retired','self$-$ employed','services','student','technician','unemployed','unknown'$)$

$3-$ marital : marital status $($categorical: 'divorced','married','single','unknown'; note: 'divorced' means divorced or widowed$)$

$4-$ education $($categorical: 'basic$\mathrm{.}4$y$\text{'},$'basic$\mathrm{.}6$y$\text{'},$'basic$\mathrm{.}9$y$\text{'},$'high.school','illiterate','professional.course','university.degree','unknown'$)$

$5-$ default: has credit in default? $($categorical: $\text{'}$no$\text{'},$'yes','unknown'$)$

$6-$ housing: has housing loan? $($categorical: $\text{'}$no$\text{'},$'yes','unknown'$)$

$7-$ loan: has personal loan? $($categorical: $\text{'}$no$\text{'},$'yes','unknown'$)$

# related with the last contact of the current campaign:

$8-$ contact: contact communication type $($categorical: 'cellular','telephone'$)$

$9-$ month: last contact month of year $($categorical: 'jan', 'feb', 'mar', $...,$ 'nov', 'dec'$)$

$10-$ day$\_$of$\_$week: last contact day of the week $($categorical: 'mon','tue','wed','thu','fri'$)$

$11-$ duration: last contact duration, in seconds $($numeric$).$ Important note: this attribute highly affects the output target $($e$.$g$.,$ if duration$=0$ then y$=\text{'}$no$\text{'}).$ Yet, the duration is not known before a call is performed. Also, after the end of the call y is obviously known. Thus, this input should only be included for benchmark purposes and should be discarded if the intention is to have a realistic predictive model.

# other attributes:

$12-$ campaign: number of contacts performed during this campaign and for this client $($numeric$,$ includes last contact$)$

$13-$ pdays: number of days that passed by after the client was last contacted from a previous campaign $($numeric; $999$ means client was not previously contacted$)$

$14-$ previous: number of contacts performed before this campaign and for this client $($numeric$)$

$15-$ poutcome: outcome of the previous marketing campaign $($categorical: 'failure','nonexistent','success'$)$

# social and economic context attributes

$16-$ emp.var.rate: employment variation rate $-$ quarterly indicator $($numeric$)$

$17-$ cons.price.idx: consumer price index $-$ monthly indicator $($numeric$)$

$18-$ cons.conf.idx: consumer confidence index $-$ monthly indicator $($numeric$)$

$19-$ euribor$3$m: euribor $3$ month rate $-$ daily indicator $($numeric$)$

$20-$ nr$.$employed: number of employees $-$ quarterly indicator $($numeric$)$

Output variable $($desired target$)$:

$21-$ y $-$ has the client subscribed a term deposit? $($binary: 'yes',$\text{'}$no$\text{'})$

You are going to build predictive models for the prediction of the ouput whether a given client will subscrive a term deposit or not. You will use data in $$bank$-$additional$-$full.csv$$ file. Some attributes have $$unknown$$ or $$nonexistent$$ categories. Don$$t bother to clean this data. You can consider them as a category in that attribute. Note that you may not get the exact results with results given in the assignment. Slightly different results are fine.

You are asked to perform the following tasks: $1.$ Build a Random Forest model.

Follow the instructions given when building the model.

Using gridsearch try to find the best score and combination of the following hyperparameters: Number of estimators: $10,50,100,250,500,1000$

max$\_$depth: $50,150,250$

min$\_$samples$\_$split: $2,3$

min$\_$samples$\_$leaf: $1,2,3$

For cross$-$validation in grid search use a cross validation strategy as $3-$fold cross$-$validation with $3$ repetitions.

Report the hyperparameter set yielding the best score. For scoring use AUC score. $2.$ Build a neural network model:

Follow the instructions given when building the model.

First, scale your input data so that it has zero mean and one standart deviation. This is important because neural network models are sensitive to input scaling.

Then using gridsearch try to find the best score and combination of the following hyperparameters: hidden$\_$layer$\_$sizes: $(10,10,10),(10,10,10,10),(10,10,10,10,10),(10,10,10,10,10,10)$

alpha: $0\mathrm{.}00001,0\mathrm{.}0001,0\mathrm{.}001,0\mathrm{.}01,0\mathrm{.}1$

In grid search fit the use AUC score as scoring. For cross$-$validation in grid search use a cross validation strategy as $3-$fold cross$-$validation with $3$ repetitions.

Report the hyperparameter set yielding the best score.