Solve the following problem in python:

Problem $4$: Decision Trees From a Relation $($or Table$)$

One of the most successful techniques in $Al$ is building a function $f$:$xY.$ We are given

data in the torm

data $=\{\begin{array}{ccc}(x_{00},& x_{01},\text{d}\text{o}\text{t}\text{s}\text{,}{x}_{0m}),& y_0\\ (x_{10},& x_{11},\text{d}\text{o}\text{t}\text{s}\text{,}{x}_{1m}),& y_1\\ \text{d}\text{o}\text{t}\text{s}\\ (x_{n0},& x_{n1},\text{d}\text{o}\text{t}\text{s}\text{,}{x}_{n1}),& u_n)(3\end{array}$

as a relation: where TID is the tuple ID and each row is called a tuple. $x$ collectively are known

as the features and $Y$ the label. In database parlances $x,Y$ are known as attributes. Given the

relation $($or data$)$ we build a decision tree that's composed from a series of teatures and values.

The function we'll be bullding in this problem is for is this data:

A Person experienced $90$ days past due delinquency or worse as $\frac{0}{1}($Boolean$)$

B Total balance on credit cards and personal lines of credit except real estate and no install$-$

ment debt like car loans divided by the sum of credit limits as percentage $($float$)$

C Age of borrower in years as integer

D Number of times borrower has been $30-59$ days past due but no worse in the last $2$ years

as integer

E Monthly debt payments, alimony,living costs divided by monthy gross income as percent$-$

age $($float$)$

F Monthly income as float

G Number of Open loans $($installment like car loan or mortgage$)$ and Lines of credit $($e$.$g$.$

credit cards$)$ as integer

$H$ Number of times borrower has been $90$ days or more past due integer as

I Number of mortgage and real estate loans including home equity lines of credit as integer

$J$ Number of times borrower has been $60-89$ days past due but no worse in the last $2$ years

as integer

K Number of dependents in family excluding themselves $($spouse$,$ children etc.$)$ as integer

Here are three tuples: So$,$ the first tupe indicates the person was delinquent of at least a

month $($Feature A$\text{'}$s definition$),$ but the other two were not. We want to build a function $f$ :

BCDEFGHIJK $$ A so that when we see a person's history:

$0\mathrm{.}025892469,38,0,74\mathrm{.}5,1,13,0,0,0,2$

$0\mathrm{.}762158561,44,4,0\mathrm{.}224465502,21000,5,0,1,0,3$

we can predict if there will be a likely delinquency at least a month. A decision tree will build

a structure that will try to solve this problem. We will have spoken about this in lecture, but

you should visit: bttps:$//$scikit$-$learn.org$/$stable$/$rodules$/$tree$.$ html and run the example

with the iris data with which you are familiar. The tree can be visualized as shown in Fig. $4$:

The tollowing code:

which means the first person will, according to the data, be deliquent, and the second one will

not. Your function will take a string $($name of the file$),$ and use pandas to do the tollowing:

Clean the data by removing rows that have $Na($not a number$).$ We covered this in lecture,

but as hint, look for dropna.

Separate the table into two data frames: B$-$K $($which are the features$)$ and A which is the

label or target. You build the tree as shown in the url.

Return the the fitted tree model.

import matplotlib.pyplot as plt

import numpy as np

import sqlite$3$

import random as rn

import matplotlib

import pandas as pd

from sklearn import tree

#problem $4$

#input file name

#output fitted tree model from skicit learn

def build$\_$tree$\_$model$($credit$\_$data$)$: