The “h2o.kmeans” Function in R

• Package: h2o

• Purpose: Clustering

• General class: Algorithm

• Required argument(s):

• training_frame: The numeric data frame or H2OFrame containing the data to be clustered.

• Notable optional arguments:

• k: The number of clusters. The default is 1.

• seed: Seed for random initialization of centroids. The default is -1 (time-based random seed).

• max_iterations: The maximum number of iterations. The default is 10.

• Example:

• # Load necessary package
  library(h2o)
  library(ggplot2)
  
  # Initialize H2O
  h2o.init()
  
  # Load a sample dataset
  data <- as.h2o(iris)
  
  # Perform K-means clustering with 3 clusters
  model <- h2o.kmeans(data, k = 3)
  
  # View the cluster centers
  print(model@model$centers)
  
  # Plot the clustering
  centers = model@model$centers
  names(centers) <- c("Centroid",paste("PC",1:(ncol(centers)-2),sep = ""),"Species")
  
  ggplot() +
  geom_point(data = iris, aes(x = Sepal.Length, y = Sepal.Width, color = Species)) +
  geom_point(data = centers, aes(x = PC1, y = PC2), color = "black", size = 3, shape = 17) +
  labs(title = "K-means Clustering", x = "PC1", y = "PC2")

• Explanation: This example demonstrates how to perform K-means clustering on the Iris dataset using the h2o.kmeans function from the h2o package. The k parameter is set to 3 to specify the number of clusters. The cluster centers are then printed and plotted to examine the resulting clusters.