ISE$-291$: Homework $04$

Problem A $[100$ Marks$]$: Solve all the questions using Python. Use Pandas, Seaborn, Sklearn, etc.,

libraries for all the analysis. Consider the data given in Excel file HW$4\_$DataA. Consider the following

data description:

Table $1.$ Data description

Do the following tasks $($in exact sequence$)$ using the $"$HW$4\_$DataA" data:

A$-1.[5$ marks$]$: Read and display the data given in HW$4\_$DataA. Describe both the numeric and

categorical attributes. Refer to Table $1$ for the data description.

A$-2.[10$ marks: $2\mathrm{.}5$each : Do the necessary pre$-$processing. In specific do the following:

a$.$ Normalize the numeric attributes using min$-$max normalization scheme.

b$.$ Perform ordinal $($label$)$ encoding for ordinal attributes $($BP$,$ and Cholestrol$).$ Use dictionary

for the ordinal encoding.

c$.$ Perform one hot encoding for the categorical attribute $($Sex$)$

d$.$ Perorfm label encoding for the class $($drug$).$

A$-3.$ marks: $2\mathrm{.}5$ each$]$:

a$.$ Split the dataset into training and testing sets using train$\_$test$\_$split function with $75\%$ for

training and $25\%$ for training using random state $=10.$

b$.$ Build a decision tree classifier for predicting the class label. Fit the classifier using the

training dataset. Set random state to $100,$ criterion to entropy, and splitter to best.

c$.$ Draw the decision tree using scikit$-$learn $($sklearn$)$