Skip to contents

Perform k selection

Source: R/7_select_k.R
select_k.Rd

select_k creates the elbow plot and silhouette width plot for assistance with selection of k

Usage

select_k(
  kmeans_data,
  standardise = TRUE,
  Max.k = 15,
  v_line = NULL,
  calc_gap = FALSE,
  plot_gap = FALSE,
  output = FALSE,
  output_folder = NULL,
  verbose = FALSE
)

Arguments

kmeans_data

Data frame containing the combined PC scores and depth statistics to perform k-means on. Output from the 'combine_data()' function.

standardise

TRUE or FALSE. Whether or not to standardise the data. Defaults to TRUE.

Max.k

Numerical. Maximum value of k to try. Defaults to 15.

v_line

Numerical. Option to add a vertical line to plot at a specific value of k. Defaults to NULL.

calc_gap

TRUE or FALSE. Whether or not to calculate the gap statistic. Defaults to FALSE

plot_gap

TRUE or FALSE. Whether or not to plot the gap statistic. Defaults to FALSE.

output

Logical. If TRUE, output is saved to output_folder. Defaults to FALSE.

output_folder

Output folder path. If output = TRUE, output_folder must be provided. Defaults to NULL.

verbose

Logical. If TRUE, progress messages are shown. Defaults to FALSE.

Value

A 'ggplot' class object and creates a figure containing both the within-cluster sum of squares plot (elbow) and the average silhouette width plot for 1 to 'Max.k' clusters.

Details

This function relies on random initialisation in k-means clustering. For reproducible results, users may wish to set a random seed prior to calling this function using set.seed().

Examples

# Set file path
filepath <- system.file("extdata", package = "FishDiveR")

# Load kmeans_data
kmeans_data <- readRDS(file.path(filepath, "data/5_k-means/combined_stats.rds"))

# Run select_k function
selecting_k <- select_k(
  kmeans_data = kmeans_data,
  standardise = TRUE,
  Max.k = 8,
  v_line = 4,
  calc_gap = FALSE,
  plot_gap = FALSE,
  output = TRUE,
  output_folder = tempdir(),
  verbose = TRUE
)
#> Standardising k-means input.
#> K = 2 Average silhouette width = 0.499
#> K = 3 Average silhouette width = 0.586
#> K = 4 Average silhouette width = 0.642
#> K = 5 Average silhouette width = 0.518
#> K = 6 Average silhouette width = 0.392
#> K = 7 Average silhouette width = 0.243
#> K = 8 Average silhouette width = 0.154
#> Single tag output file: /tmp/RtmpTKWEE7/data/5_k-means/Select_k.3_PCs.png