rf-regression-R.qmd

---
title: "Random Forest Regression with R"
format: html
typora-copy-images-to: images
---

```{r}
# Load required libraries
library(randomForest)
library(ggplot2)

# Set seed for reproducibility
set.seed(123)

data <- mtcars
```

# Split data into training and testing sets


```{r}
train_indices <- sample(1:nrow(data), 0.7 * nrow(data))
train_data <- data[train_indices, ]
test_data <- data[-train_indices, ]
```

# Train Random Forest model and make predictions


```{r}
rf_model <- randomForest(mpg ~ ., data = train_data, ntree = 500)

predictions <- predict(rf_model, newdata = test_data)

```

# Calculate RMSE

```{r}
rmse <- sqrt(mean((test_data$mpg - predictions)^2))
cat("Root Mean Square Error:", rmse, "\n")

```

# Plot actual vs predicted values

```{r}
ggplot(data.frame(actual = test_data$mpg, predicted = predictions), aes(x = actual, y = predicted)) +
  geom_point(alpha = 0.5) +
  geom_abline(intercept = 0, slope = 1, color = "red", linetype = "dashed") +
  labs(x = "Actual Values", y = "Predicted Values", title = "Random Forest Regression: Actual vs Predicted")
```

# Print feature importance

```{r}
importance <- importance(rf_model)
print(importance)
```

# Plot feature importance

```{r}
importance_df <- data.frame(feature = rownames(importance), importance = importance[, 1])
ggplot(importance_df, aes(x = reorder(feature, importance), y = importance)) +
  geom_bar(stat = "identity") +
  coord_flip() +
  labs(x = "Features", y = "Importance", title = "Random Forest: Feature Importance")

```

alternative at https://hackernoon.com/random-forest-regression-in-r-code-and-interpretation