I am trying to make a bayesian network in R$.$ I am stating summary of my data and code in R$.$ but the result of my Bayesian network is seems weird and incorrect.

Please send me corrected code here.

summary$($data$)$

Age Band of Driver Sex of Driver Educational Level Driving Experience

$18-30$ :$4209$ Female : $696$ Above high school : $3575-10$yr :$3322$

$31-50$ :$4033$ Male :$11285$ Elementary school :$21392-5$yr :$2567$

Over $51$ :$1561$ Unknown: $177$ High school :$1095$ Above $10$yr:$2233$

Under $18$: $813$ Illiterate : $451-2$yr :$1735$

Unknown :$1542$ Junior high school:$7522$ Below $1$yr :$1335$

Unknown : $827$ Unknown : $816$

Writing & reading : $173($Other$)$ : $150$

Types of Junction Road Surface Type Light Conditions

Y Shape :$4490$ Asphalt roads :$11148$ Darkness $-$ lights lit :$3220$

No junction:$3766$ Asphalt roads with some distress: $80$ Darkness $-$ lights unlit: $40$

Crossing :$2165$ Earth roads : $351$ Darkness $-$ no lighting : $187$

Unknown :$1059$ Gravel roads : $242$ Daylight :$8711$

Other : $443$ Other : $166$

O Shape : $163$ Unknown : $171$

$($Other$)$ : $72$

Weather Conditions Type of Collision Vehicle Movement

Normal :$9928$ Vehicle with vehicle collision :$8670$ Going straight :$8059$

Raining:$1308$ Collision with roadside objects:$1762$ Other :$1316$

Other : $296$ Collision with pedestrians : $874$ Moving Backward: $973$

Unknown: $292$ Rollover : $393$ Reversing : $556$

Cloudy : $125$ Collision with animals : $169$ Turnover : $482$

Windy : $98$ Unknown : $167$ Getting off : $334$

$($Other$)$: $111($Other$)$ : $123($Other$)$ : $438$

Cause of Accident Accident Severity

ChangeLane :$3242$ Light : $1743$

NoDistance :$2243$ Serious:$10415$

NoPriority :$1910$

DriveHighOrCareless:$1552$

Other :$1458$

MovingBackward :$1111$

$($Other$)$ : $642$

Code: library$($readr$)$

library$($bnlearn$)$

library$($gRain$)$

library$($ggplot$2)$

library$($reshape$2)$

library$($forcats$)$

library$($dplyr$)$

data $<-$ read.csv$("$F:$/2$nd Semester$/$Data Processing$/$Bayesian$/$Dataset$/$archive $(2)/$cleaned$.$csv$")$

data $=$ lapply$($data$,$ as$.$factor$)$

data $=$ data.frame$($data$)$

levels$\_$Movement $=$ c$("$Enteringajunction$",$ "Getting off", "Going straight",

"Moving Backward","Other","Overtaking","Parked","Reversing"

$,$"Stopping","Turnover","U$-$Turn","Unknown","Waiting to go$")$

levels$($data$Vehicle$\_$movement$)=$ c$("$Entering a junction", "Getting off", "Going straight",

"Moving Backward", "Other", "Overtaking","Parked","Reversing"

$,$"Stopping","Turnover","U$-$Turn","Other","Waiting to go$")$

levels$($data$Cause$\_$of$\_$accident$)=$ c$("$ChangeLane$",$ "ChangeLane", "DriveHighOrCareless",

"DriveHighOrCareless", "DriveLeft", "DriveAlcohol",

"DriveAlcohol", "Other", "Other","MovingBackward",

"NoDistance", "NoPriority","NoPriority", "Other","Other",

"Other","Other","Other","Other","Other"$)$

# Convert the "Accident$\_$severity" column to character

data$Accident$\_$severity $=$ as$.$character$($data$Accident$\_$severity$)$

# Convert the "Accident$\_$severity" column to numeric

data$Accident$\_$severity $=$ as$.$numeric$($data$Accident$\_$severity$)$

# Now, use the cut$()$ function

data$Accident$\_$severity $=$ cut$($data$Accident$\_$severity, breaks $=$ c$(0,1,2),$ labels $=$ c$("$Light$",$ "Serious"$))$

data $=$ subset$($data$,$ select $=-$c$($Vehicle$\_$driver$\_$relation,Lanes$\_$or$\_$Medians,Pedestrian$\_$movement$))$

data $<-$ na$.$omit$($data$)$

data $<-$ data $\%>\%$ mutate$\_$if$($is$.$numeric, ~ifelse$($is$.$na$(.),$ mean$(.,$ na$.$rm $=$ TRUE$),.))$

column$\_$names$=$ c$("$Age Band of Driver", "Sex of Driver", "Educational Level", "Driving Experience","Types of Junction",

"Road Surface Type", "Light Conditions", "Weather Conditions", "Type of Collision","Vehicle Movement", "Cause of Accident", "Accident Severity"$)$

colnames$($data $)=$ column$\_$names

summary$($data$)$

tabu$($data$)$

mmhc$($data$)$

h$2$pc$($data$)$

# Create a Bayesian network structure using the 'hill climbing' algorithm

network $<-$ hc$($data$)$