### in case we haven't installed the necessary libraries
# install.packages("rpart")
# install.packages("rpart.plot")

##########
#Example 1 (on iris)
##########

# Load the necessary libraries 
library(rpart) 
library(rpart.plot) 

# Load and inspect the iris dataset 
data(iris) 
pairs(iris)

# Build a classification decision tree 
iris_tree <- rpart::rpart(Species ~ ., data = iris, method = "class") 

# Plot the decision tree 
rpart.plot(iris_tree, main = "Decision Tree for Iris Species") 

# Make predictions on the dataset 
predictions <- predict(iris_tree, iris, type = "class") 


# Generate confusion matrix 
conf_mat <- table(Predicted = predictions, Actual = iris$Species) 

# Display the confusion matrix 
print(conf_mat) 

# print model's summary
summary(iris_tree)


##########
#Example 2 (on mtcars)
##########
# Build a regression tree to predict mpg 
mtcars_tree <- rpart(mpg ~ wt + hp + cyl, data = mtcars, method = "anova") 

# Plot the regression tree 
rpart.plot(mtcars_tree, main = "Regression Tree for mpg") 

# inspect the tree-inferred rules
rpart.rules(mtcars_tree)

# plot complexity parameter for the fit used in rpart
plotcp(mtcars_tree)

# Make predictions 
predicted_mpg <- predict(mtcars_tree, mtcars) 

# Compare actual vs predicted 
head(cbind(Actual = mtcars$mpg, Predicted = predicted_mpg)) 



##########
#Example 3 (plotting the trees with base R)
##########

# Basic plot of the tree structure 
plot(iris_tree, uniform=TRUE, main="Decision Tree Structure") 
text(iris_tree, use.n=TRUE, all=TRUE, cex=0.8)