Principal Component Analysis (PCA) and Visualization

# File Name: principal_component_analysis.R

# Author: Gerard King - www.gerardking.dev

# Title: Principal Component Analysis (PCA) and Visualization

# Description: This R program generates a synthetic dataset, applies PCA to reduce the data's

# dimensionality, and visualizes the first two principal components. PCA is widely used for 

# feature extraction and data visualization in high-dimensional datasets.

# Use Cases:

# - Reducing the dimensionality of high-dimensional data

# - Visualizing high-dimensional datasets in lower dimensions

# - Identifying patterns or structures in complex data

# Audience:

# - Data scientists and machine learning practitioners

# - Students learning dimensionality reduction techniques

# - Researchers analyzing multivariate datasets

# Blue Team Uses:

# - Reducing system performance metrics (e.g., CPU, memory, network) to simpler features

# - Visualizing high-dimensional logs or system data for easier monitoring and anomaly detection

# - Identifying patterns in complex security data

# Red Team Uses:

# - Reducing attack simulation data to identify patterns and critical features

# - Using PCA to visualize attack data in lower dimensions for easier analysis

# - Identifying the most critical variables in simulated attacks to help in prioritizing security measures

# Current Date: 2025-03-06

# Load necessary library

library(ggplot2)

# Set the seed for reproducibility

set.seed(303)

# Generate synthetic multivariate data with 5 features

data <- data.frame(

  var1 = rnorm(100),

  var2 = rnorm(100),

  var3 = rnorm(100),

  var4 = rnorm(100),

  var5 = rnorm(100)

)

# Apply Principal Component Analysis (PCA)

pca_result <- prcomp(data, center = TRUE, scale. = TRUE)

# Summary of the PCA to see variance explained by each principal component

cat("Summary of PCA:\n")

print(summary(pca_result))

# Create a data frame with the first two principal components

pca_data <- as.data.frame(pca_result$x[, 1:2])

# Plot the first two principal components

ggplot(pca_data, aes(x = PC1, y = PC2)) +

  geom_point(color = "blue", size = 3) +

  labs(title = "PCA: First Two Principal Components", 

       x = "Principal Component 1", y = "Principal Component 2") +

  theme_minimal()

# Display the explained variance (proportion of variance) for the first two PCs

cat("\nProportion of Variance Explained by the First Two Principal Components:\n")

print(summary(pca_result)$importance[2, 1:2])

# Print the current date for reference

cat("Date of execution:", Sys.Date(), "\n")