sample.r

# Importing dataset
dataset <- read.csv('ASD.csv')

# Taking care of missing data
# Fixing ethnicity
dataset[dataset == "?"] <- NA
summary(dataset)

#install.packages('DescTools')
library(DescTools)

DetermineEthnicity = function(x){
  eth = Mode(dataset[dataset$contry_of_res == x, ]$ethnicity)
  if (length(eth) == 1){
    return (eth)
  }else if(length(eth) == 12){
    return(Mode(dataset$ethnicity))
  }else{
    return(eth[1])
  }
}

for(r in 1:nrow(dataset)){
  if(is.na(dataset[r, ]$ethnicity)){
    dataset[r, ]$ethnicity = DetermineEthnicity(dataset[r, ]$contry_of_res)
  }
}

# Fixing relations
self = dataset[1, "relation"]
for(r in 1:nrow(dataset)){
  if(is.na(dataset[r, ]$relation)){
    dataset[r, ]$relation = self
  }
}

# Converting Age to numeric type
dataset$age = as.numeric(as.character(dataset$age))

# Dividing into training and test set
#install.packages('caTools')
library(caTools)
split = sample.split(dataset$Class.ASD, SplitRatio = 7/10)
training_set = subset(dataset, split == TRUE)
test_set = subset(dataset, split == FALSE)
X_test = test_set[, 1:21]
y_test = test_set[, 22]

# Fitting decision tree regression to the dataset
#install.packages('rpart')
library(rpart)
regressor = rpart(formula = Class.ASD ~ .,
                  data = training_set)

# Predicting result and checking accuracy
y_pred = predict(regressor, newdata = X_test, type = "class")

print(mean(y_test == y_pred))