Introduction to Machine Learning in R

Machine learning enables computers to learn patterns from data and make predictions. R provides multiple libraries for machine learning, including caret and randomForest.

Loading the Caret Package

The caret package simplifies machine learning workflows in R.

library(caret)
data(iris)

Splitting Data into Train and Test Sets

Splitting data ensures proper model evaluation by training and testing separately.

set.seed(123)
trainIndex <- createDataPartition(iris$Species, p=0.8, list=FALSE)
trainData <- iris[trainIndex,]
testData <- iris[-trainIndex,]

Building a Decision Tree Model

Decision trees classify data by splitting features at decision points.

library(rpart)
model <- rpart(Species ~ ., data=trainData)
plot(model)
text(model)

Training a Random Forest Model

Random forests improve decision trees by combining multiple models.

library(randomForest)
rf_model <- randomForest(Species ~ ., data=trainData)
print(rf_model)

Making Predictions

Use predict() to make predictions on new data.

predictions <- predict(rf_model, newdata=testData)

Evaluating Model Accuracy

Accuracy measures how well the model performs on test data.

confusionMatrix(predictions, testData$Species)

Hyperparameter Tuning

Use train() to optimize model parameters for better performance.

tuned_model <- train(Species ~ ., data=trainData, method='rf', tuneLength=3)

Summary

In this lesson, you learned:

• How to split data into training and testing sets.
• How to build decision tree and random forest models.
• How to evaluate model performance and tune hyperparameters.