$5|$ P a g e

Task

First, copy the code below to a R script. Enter your student ID into the command set.seed$(.)$

and run the whole code. The code will create a sub$-$sample that is unique to you.

Use the str$(.)$ command to check that the data type for each feature is correctly specified.

Address the issue if this is not the case.

You are to clean and perform basic data analysis on the relevant features in mydata, and as

well as principal component analysis $($PCA$)$ on the continuous variables. This is to be done

using $$R$.$ You will report on your findings.

Part $1$ Exploratory Data Analysis and Data Cleaning

$($i$)$ For each of your categorical or binary variables, determine the number $(\%)$ of

instances for each of their categories and summarise them in a table as follows.

State all percentages in $1$ decimal places.

Categorical Feature Category N $(\%)$

Feature $1$ Category $110(10\mathrm{.}0\%)$

Category $230(30\mathrm{.}0\%)$

Category $350(50\mathrm{.}0\%)$

Missing $10(10\mathrm{.}0\%)$

Feature $2($Binary$)$ YES $75(75\mathrm{.}0\%)$

NO $25(25\mathrm{.}0\%)$

Missing $0(0\mathrm{.}0\%)$

$.........$

Feature k Category $125(25\mathrm{.}0\%)$

Category $225(25\mathrm{.}0\%)$

Category $315(15\mathrm{.}0\%)$

Category $430(30\mathrm{.}0\%)$

Missing $5(5\mathrm{.}0\%)$

# You may need to change$/$include the path of your working directory

dat $<-$ read.csv$("$HealthCareData$\_2024.$csv$",$ stringsAsFactors $=$ TRUE$)$

# Separate samples of normal and malicious events

dat.class$0<-$ dat $\%>\%$ filter$($Classification $==$ "Normal"$)$ # normal

dat.class$1<-$ dat $\%>\%$ filter$($Classification $==$ "Malicious"$)$ # malicious

# Randomly select $400$ samples from each class, then combine them to form a working dataset

set.seed$($Enter your student ID here$)$

rand.class$0<-$ dat.class$0[$sample$(1$:nrow$($dat$.$class$0),$ size $=400,$ replace $=$ FALSE$),]$

rand.class$1<-$ dat.class$1[$sample$(1$:nrow$($dat$.$class$1),$ size $=400,$ replace $=$ FALSE$),]$

# Your sub$-$sample of $800$ observations

mydata $<-$ rbind$($rand$.$class$0,$ rand.class$1)$

dim$($mydata$)$ # Check the dimension of your sub$-$sample